key: cord-0823671-401l7jbn
authors: Chamieh, A.; Afif, C.; El-Hajj, G.; Zmerli, O.; Djaffar-Jureidini, I.; Raad, R. A.; Ashou, R.; Juvelekian, G.; Rolain, J.-M.; Azar, E.
title: Viral Dynamics Matter in COVID-19 Pneumonia: the success of early treatment with hydroxychloroquine and azithromycin in Lebanon.
date: 2020-06-02
journal: nan
DOI: 10.1101/2020.05.28.20114835
sha: 500beca0258a01b21a00875e06206149f514287f
doc_id: 823671
cord_uid: 401l7jbn

Background: We share our experience in COVID-19 pneumonia management at Saint George Hospital University Medical Center (SGHUMC) in Lebanon. In the absence of a standard of care, early diagnosis and opt-in therapy with Hydroxychloroquine and Azithromycin were offered. Methods: We reviewed records of COVID-19 pneumonia patients from March 16-April 26 2020. Based on NEWS score, we stratified patients as A: low B: medium, and C: high clinical severity and obtained pharmacotherapy data. Chest-CT-severity-score (CTSS) was used. We defined clinical cure as resolution of symptoms and biomarkers and virologic cure as a PCR above 35 cycles(Ct). Results: We recorded 21 COVID-19 pneumonia patients of whom 19 opted for treatment. Clinical symptoms and laboratory markers at presentation did not significantly correlate with severity. Lower initial viral load significantly correlated with lower levels of clinical and radiological severity (p=0.038). Virologic cure, Ct>35, by day 10, was only 33% in high severity significantly less than categories A and B. We observed 100% clinical cure at day 10 in Category-A, 67% in B, and 33% in C(p<0.05). Patients with the lowest severity had the fastest virologic cure in a mean of 5.8 days from diagnosis, shortest hospitalization and earlier radiological improvement(p<0.005). Ultimately, 18 patients were discharged home in good condition and one remains in the ICU. Conclusion: Viral dynamics matter in COVID-19 pneumonia. An early control of replication may be crucial in averting complications. Early administration of Hydroxychloroquine and Azithromycin potentially explains our 94.7% success rate in treating a fairly complex cohort of COVID-19 pneumonia.

We conducted this retrospective study at SGHUMC in Beirut, Lebanon, from 112 March 16, 2020 until April 26, 2020 Patients were received via our Emergency Room and Corona Clinic or as 114 transferals requiring urgent critical care, some of which were COVID-19 confirmed. 115

We obtained exposure and contact history of all incoming patients with 116 respiratory or febrile illnesses and performed a physical exam. Nasopharyngeal and 117 oropharyngeal swabs were obtained for SARS-CoV-2 RT-qPCR. Chest CT imaging was 118 done within 3 days of diagnosis. 119

The SGHUMC response team was composed of infectious disease physicians, 120 pulmonary and critical care physicians, radiologists, infection preventionists, nursing 121 department, laboratory department, and a representative from hospital leadership. 122

Clinical and virologic cure was assessed at day 10 from diagnosis. Patients were 123 followed until complete clinical and virologic clearance, defined as the overall outcome. 124

In our analysis, we included all patients that were positive for SARS-CoV-2 126 by RT-qPCR with findings of pneumonia on chest CT scan and had received 127

Hydroxychloroquine and azithromycin for at least 72 hours. 128

We retrieved data from the electronic database of the COVID-19 clinic and 130 emergency care service. We recorded patient age, gender, history of cardiovascular 131 disease, hypertension, chronic lung conditions, diabetes mellitus, malignancy, and any 132 immunosuppressive therapy. We also recorded possible exposure history and 133 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 2, 2020. We used the National Early Warning Score (NEWS) and chest CT scan findings to 145 assess acute clinical severity on admission. 10 The NEWS is based on physiologic 146 parameters that include respiratory and heart rate, blood pressure, oxygen saturation, use 147 of supplemental oxygen, and state of consciousness. A NEWS score of 0-4, 5-6, and 7-15 148 respectively corresponded to low, medium, and high scores. For the sake of data analysis, 149 we then divided our cohort into categories A, B, and C according to low, medium, and 150 high NEWS score. 151

We define clinical cure as the resolution of symptoms, clinical stabilization, and 152 normalization of inflammatory markers. 153

Patients were managed as inpatients or followed as outpatients depending on 154 their clinical severity. Outpatients were discharged with strict instructions and 155 education, and were called for follow up every 2 days. 156 3.5. PCR Assay 157 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 2, 2020. Molbiol, Berlin, Germany) as previously described. 11, 12 We used the cycle number 161 threshold values (Ct) as an indicator of the number of copies of SARS-CoV-2 present 162 in our samples. We define virologic cure as 2 consecutive SARS-CoV-2 RT-qPCR 163 with Ct>35, as previously described. 13 164

We considered a patient with virologic cure by day 10 of diagnosis a fast 165 responder and a slow responder if more than 10 days. All patients were followed with 166 serial RT-PCR till negative, beyond limit of detection of the test, as part of SGHUMC 167 infection control strategy. 168

We also tested for other pathogens using Biofire® FilmArray® Respiratoy 2 169

Panel (BioMerieux, France) following manufacturer's instructions. COVID-19 pneumonia. The scoring system was adapted from similar scoring systems 182 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint 13 used in recent COVID-19 literature. 14,15 A score of 0, 1, 2, or 3 (corresponding to 183 involvement of 0%, 0-25%, 25-50%, and >50% respectively) was assigned for each 184 lobe of the 5 lung lobes with a maximum total score of 15. Therefore, if a patient has 185 a CTSS of 15, it indicates that there is more than 50% involvement of the lung. 186

Interval improvement and resolution is a decrease in CTSS and/or subjective decrease 187 in ground-glass opacities density. 188

All images were anonymized then analyzed by two experienced radiologists 189 (R.A. and R.R.) independently. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. There was no distinct pattern for progression of laboratory markers. However, when 258 peak values are reached, it took 4-6 days to return to baseline. This was independent of 259 initial clinical severity. Table 1 100% clinical cure at day 10 in low clinical severity, 67% in medium severity, and 266 33% in high severity COVID-19. This trend was statistically significant (p=0.031). 267 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. Figure 1A) . At the end of the study, SARS-CoV-2 RT-PCR results were available for all 288 patients. In category A, eight of ten patients achieved virologic cure, beyond the limit 289 of detection of our test at mean of 14.8±7.7 days. In category B, all patients achieved 290 virologic cure beyond limit of detection of our test and clinical cure at a mean of 291 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint 16±5 days. As for category C, only 2 out of the 3 patients achieved virologic cure 292 beyond limit of detection of our test on days 21 and 24. (Figure 2 ) 293 Patients with lower clinical severity reached virologic cure significantly faster 294 than those of higher severity (p=0.042). 295 4.3. Radiologic Findings: 296

There was a strong, positive correlation between the two readers R.A. and 297 R.R., which was statistically significant (rs(36) = 0.958, p < 0.001). 298

All patients had a chest CT scan done within 3 days of COVID-19 diagnosis. 299

To note, the median time from symptom onset to imaging was 3 days for categories A 300 and C and 8 days for category B. (p = 0.842). 301

On initial CT, there was a statistically significant difference in chest CT 302 Severity Score(CTSS) between the 3 categories (p = 0.012) ( Figure 1B) . Patients of 303 category B showed the highest median CTSS of 10/15 compared to A and C (4/15 (p 304 = 0.003) and 6/15 (p = 0.192), respectively). (Figures 3,4) . 305

Only 15 patients underwent follow-up chest CT scans at a median of 14 days 306 (range 10-26 days). Thirteen (87%) patients showed significant interval improvement 307 of which two patients showed complete resolution (p = 0.031). ( Figure 5 ) 308

The remaining two patients with worsening features on second CT showed 309 significant improvement on the subsequent third CT scan (Figure 6 ), obtained on day 310 20 and 29 respectively. 311

The median follow-up interval from time of diagnosis to radiological 312 improvement for the 15 patients was: 14 (10-24), 12 (10-29), and 20 (14-26) days for 313 categories A, B, and C respectively. 314 315 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint

Combination of hydroxychloroquine plus azithromycin as potential treatment for 411

COVID 19 patients: pharmacology, safety profile, drug interactions and management 412 of toxicity

Early treatment 414 of COVID-19 patients with hydroxychloroquine and azithromycin: A retrospective 415 analysis of 1061 cases in Marseille, France

Recycling of chloroquine and its hydroxyl analogue to 418 face bacterial, fungal and viral infections in the 21st century

Projection of Optimized Dosing Design of Hydroxychloroquine for the Treatment of 422

Severe Acute Respiratory Syndrome Main point : Hydroxychloroquine was found to 423 be more potent than chloroquine at inhibiting SARS-CoV-2 in vit

FDA approved drugs with 426 broad anti-coronaviral activity inhibit SARS-CoV-2 in vitro

New insights on the antiviral effects of 428 chloroquine against coronavirus: what to expect for COVID-19?

Clinical trials on drug repositioning for COVID-19

At SGHUMC, 78% of patients diagnosed with COVID-19 had pneumonia on chest CT 317 scan. Symptoms such as shortness of breath and dyspnea were not reliably predictive of 318 disease severity. Our low threshold to perform chest CT scan was crucial in early diagnosis. 319In our cohort, an initially low viral load was associated with less severe clinical and 320 radiological findings, and faster viral clearance. All patients in category A achieved clinical 321 cure in 10 days and had the fastest virologic cure in a mean 5.8 days from diagnosis. In 322 contrast, only 56% of patients of higher clinical severity achieved clinical cure in 10 days, 323 had longer hospital stays, and a more prolonged viral shedding. A recent study on 1061 324 patients with COVID-19 at IHU-Mediterranee Infection in Marseille, France showed that 325 patients with higher initial viral load were less likely to have a low clinical severity and more 326 likely to have poor virologic outcome. 17 Our results are consistent with their findings 327 confirming that initial viral load and kinetics of viral clearance matter in In many viral infections, early control of viral replication is crucial for better outcomes. 329Hydroxychloroquine has an antiviral effect that is proven in-vitro and in-vivo. 18-23 We had a 330 94.7% success rate in treating a fairly complex cohort of COVID-19 pneumonia, with 18/19 331 patients discharged at baseline conditions. None of our 19 pneumonia patients had a cytokine 332 storm, and all had a rapid decline within 4 days from peak inflammatory laboratory markers. 333The latter is likely a result of virologic control and possible early attenuation of cytokine 334 storm. 7,24,25 335Chest CT scan is a helpful diagnostic tool but also serves to monitor progression. Thirteen 336 of 15 follow-up chest CT scans improved within 14 days. The remaining two showed 337 improvement in a third scan beyond 20 days. In fact, all patients improved, showing either 338 complete resolution or marked improvement in the CT findings, mainly ground-glass 339 opacities and parenchymal consolidations. A follow-up chest CT may be necessary in 340 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint 20 patients with COVID-19 to monitor for secondary organizing pneumonia and subsequent 341 fibrosis, similar to outcomes seen with SARS and MERS. 14 We will continue to monitor our 342 patients for such complications despite the current significant radiologic improvement. 343Our observational retrospective cohort is not designed to measure the direct impact of 344 hydroxychloroquine and azithromycin treatment on the outcome of COVID-19 pneumonia. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 2, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted June 2, 2020. . https://doi.org/10.1101/2020.05.28.20114835 doi: medRxiv preprint