Empirical treatment with hydroxychloroquine and azithromycin for suspected cases of COVID-19 followed-up by telemedicine

Posted by

Salient features:

This article focuses on:

  • Background
  • Aim
  • Methods
  • Study population
  • Treatment protocol
  • Clinical outcome
  • Study design and data collection
  • Statistical analysis
  • Results
  • Discussion
  • Clinical implications
  • Conclusion

Detailed summary:


  • Telemedicine has been described as a potential tool for mitigating the impact of disasters, health emergencies and providing public services.
  • “Direct screening” by telemedicine, described as the classification of patients before they reach the emergency department, promotes social isolation.
  • Telemedicine allows patients to be tracked efficiently and possibilities communication between doctors and patients 24/7 through smartphones or webcam-enabled computers.


  • It has been used as first-line drug in the treatment and prophylaxis of malaria.
  • Chloroquine has antiviral activity against several RNAviruses, such as rabies, poliovirus, Hepatitis A and C viruses, Influenza A and B , Dengue, Zika and also against coronavirus.
  • Mechanism of action includes blocking cell infection by increasing the endosomal pH and interfering with the glycosylation of the SARS-CoV2 cell receptor.
  • Hydroxychloroquine, a derivative of chloroquine, has a hydroxyl group at the end of the side chain, having pharmacokinetics similar to chloroquine, with rapid gastrointestinal absorption and renal excretion in addition to a less toxic profile.
  • Hydroxychloroquine also inhibits SARS-CoV-2 infection in vitro.
  • Adding azithromycin to hydroxychloroquine increased treatment’s effectiveness.
  • These drugs have limited toxicity and generally mild and transient side effects.
  • The use of chloroquine for more than 70 years as an antimalarial treatment reinforces its safety for acute administration.
  • Cost effective.
  • Hydroxychloroquine plus azithromycin on early stages of COVID-19 could inhibit viral replication and prevent progression to severe forms of the disease, it is rational to hypothesize that treating patients at the beginning of the viral infection could have potential benefits, possible decreasing the need for hospitalization.


To assess whether empirical prescription of hydroxychloroquine and azithromycin for patients with suspected COVID-19 is associated with less need for hospitalization


Study population:

  • Patients enrolled in the study were residents of the city of Sao Paulo, Brazil.
  • Outpatients with persistent flu-like symptoms (suspected COVID19 infection), persisting for a period equal to or greater than 2 days, were first evaluated by the telemedicine team or by the emergency department medical doctor.
  • All physicians had access to medical records of all subjects, such as clinical history, laboratory parameters, images exams and electrocardiograms.
  • Those who had no immediate need for hospitalization and no contraindications for treatment were invited to participate in the study.
  • Treatment with hydroxychloroquine associated with azithromycin was prescribed.
  • Chest computed tomography was performed according to medical judgment.
  • Definitions of the severity of lung injury according to tomographic aspects were: Mild (<25% of lung involvement), moderate (25% to 50% of lung involvement) and high (>50% of lung involvement).
  • The main hospitalization admission criteria were:
  • Worsening general condition
  • Oxygen Saturation < 90%
  • Inclusion criteria:
  • Patient over 18 years old and flu-like persisteent symptoms > 3 days, with a probable diagnosis of COVID-19 and no immediate indication for hospitalization.
  • Exclusion criteria :
  • Severe related retinopathy
  • Severe liver disease
  • Myasthenia Gravis
  • Known QT enlargement
  • Pregnant
  • Severe renal failure

Treatment protocol

  • Hydroxychloroquine 800mg on the first day and 400mg for another 6 days and azithromycin 500mg once daily for five days.

Clinical outcomes :

  • To evaluate whether the empirical prescription of Hydroxychloroquine plus Azithromycin in outpatients is associated with less need for hospitalization.
  • To evaluate the difference for hospitalization in patients treated before and after the seventh day of symptoms observation.

Study design and data collection

  • Consecutive patients with flu-like symptoms with no indication for hospitalization were included and followed up by telemedicine healthcare team.
  • All patients were informed about the efficiency of azithromycin and hydroxychloroquine in treating COVID-19, safety profile of the drugs and potential side effects
  • The consent form was electronically sent to the patient and signed on line, during telemedicine call or presently when the first evaluation was done in the emergency department.
  • Hydroxychloroquine plus azithromycin were delivered at home to all those who accepted the term and agreed to use the medication.
  • A telemedicine platform, with HIPAA compliance certified system for data security, was used for medical consultations.
  • All patients were followed daily by telemedicine consultations until the fifth day of symptoms, after that, patients were contacted twice a day until the fourteenth day of initial symptoms.
  • Breathing pattern was evaluated during videoconference.
  • All patient data were taken from electronic medical records.
  • It was defined as the treatment group, patients that accepted the treatment with hydroxychloroquine plus azithromycin. It was defined as the control group, patients that refused and did not sign the informed consent to use hydroxychloroquine and azithromycin.

Statistical analysis

  • The paired-sample t test and the unpaired-sample t test were used to compare means within the study group or between subgroups.
  • The chi-square and the Fisher exact tests were used for comparison of discrete variables.
  • The number needed to treat (NNT) was calculated by the inverse of the absolute risk reduction (ARR) expressed as a decimal.
  • Continuous variables without normal distribution were compared using the Mann-Whitney U test, and correlation between such variables using the Spearman rank test.
  • Values of p < 0.05 were considered statistically significant.
  • The statistical analysis was conducted using the statistical package SPSS 15.

Results :

  • 721 patients with flu-like symptoms were referred to telemedicine service, of these 85 patients were not followed due to difficulties in technical communication and lack of registration.
  • Telemedicine team followed 636 consecutive outpatients who had flu-like symptoms and could be monitored.
  • Of these, 224 patients refused the proposed treatment, making up the control group; 412 consented to start treatment with hydroxychloroquine and azithromycin.
  • The average time of symptom onset in which the medication was prescribed was 5.2 ± 3.1 days and the average follow up was 5.0 ± 2.7 days.
  • The mean age was 62.5± 15.5 years and 400 were female.
  • 85 patients had a diagnosis of type 2 diabetes mellitus, 168 had a history of hypertension, 49 were obese and 17 patients were smokers during the inclusion period of the study.
  • The baseline clinical characteristics were similar between groups except by a higher rate of diabetes and previous stroke in the treatment group.
  • The treatment group also had higher prevalence of flu-like symptoms than the control group, such as fever, cough, dyspnea, diarrhea, myalgia, coryza, and headache.
  • Dyspnea at baseline was more prevalent in the treatment group compared to controls.
  • Only patients with dyspnea showed greater improvement in the treatment during the follow-up.
  • Chest CT was performed in 251 subjects in the treatment group and showed that 70.1% had COVID-19 suggestive images; 150 patients had mild lung involvement, 26 moderate and none showed severe lung compromising.
  • Only 54 chest CT were performed in the control group and of those 40,7% had COVID-19 suggestive images.
  • All patients from both groups who needed hospitalization presented COVID-19 pattern at chest CT.
  • There were no serious side effects in patients treated with hydroxychloroquine plus azithromycin.
  • Two patients in the treatment group died during the follow-up; first death was due to acute coronary syndrome and second death due to metastatic cancer.
  • On the treatment group, 1.9% required hospitalization, compared to the control group, which was 5.4%.
  • Patients treated before versus after day 7 of symptoms required less hospitalization.
  • Comparing the early treatment (< 7 days of symptoms) to those without treatment (control group) the NNT was 23.

Discussion :

  • Early evaluation of suspected COVID-19 patients by telemedicine associated with empirical treatment with hydroxychloroquine and azithromycin is an important strategy that may prevent hospitalization.
  • Patients treated with hydroxychloroquine and azithromycin compared to untreated patients had 2.8-fold lower need for hospitalization.
  • In addition, need for hospitalization in patients treated before versus after the 7th day of symptoms were 1.17% and 3.2%, respectively,, that is a 2.7x lower rate of hospitalization when treatment was started earlier and 4.6x lower rate of hospitalization compared to untreated patients.
  • Patients hospitalized with severe COVID-19 have laboratory evidence of “cytokine storm” with persistent fever, elevated inflammatory markers and proinflammatory cytokines.
  • The pathophysiological rationale of this study is that starting treatment empirically, allows hydroxychloroquine and azithromycin to act in a milder phase of the disease, possibly decreasing viral replication and preventing progression to aggressive stages.
  • In a study of 1014 patients in Wuhan who underwent a polymerase chain reaction with reverse transcription (RT-PCR) and chest computed tomography to assess COVID-19, a “positive” CT for COVID-19 presented a sensitivity of 97%, using the C-reactive protein (PCR) tests as a reference
  • In this study, 60.9% of patients in the treatment group underwent chest CT, 70.1% of those had characteristics findings of COVID-19 , suggesting high probability of COVID-19 in our population.
  • The duration of viral shedding is also variable and may depend on the severity of the disease.
  • In a study with 21 patients with mild disease (without hypoxia), 90% repeatedly tested negative for viral RNA in nasopharyngeal swabs for 10 days after the onset of symptoms; the tests were positive for an extended period in patients with more severe diseases.
  • Such data emphasize that initiation of treatment cannot depend on laboratory tests alone, given the high rate of false negatives and the delay in obtaining the result.
  • Swab PCR results can take days until diagnosis, which can be crucial for clinical evolution of the infected patient.
  • Liu et al demonstrated that the “waiting period” may be critical for patients to entering high inflammatory and immune response phase.
  • Patients with suspected COVID-19 and mild symptoms who do not need emergency care should be encouraged to stay home and seek a telemedicine appointment before going to a health facility.
  • Clinical status should be assessed during videoconference by tests and evaluation of breathing pattern is decisive to classify severity and support decision-making process.
  • Therefore, telemedicine is an auspicious tool that can effectively provides patient’s care, reduce emergency unit overcrowding and promotes social isolation.

Clinical implications:

  • In the study, despite patients in the treatment group had higher 16 prevalence of diabetes, immunosuppression state and p-trend for history of stroke, the need for hospitalization was smaller.
  • Hydroxychloroquine plus azithromycin is one of the most promising alternatives to treat COVID-19.
  • Its low cost and safety profile make its reasonable to large-scale use.
  • Non-serious adverse events, such as vomiting, nausea, headache and abdominal pain, were most commonly reported and only two cases were considered serious (maculopapular rash and severe pruritus).
  • None of the patients required hospitalization.
  • Hydroxychloroquine is about 40% less toxic than Chloroquine.
  • The initial phase of the disease seems the most rational to start hydroxychloroquine and azithromycin and diagnosis of COVID-19 should not be based on swab collection alone, once it can delay initiation of treatment.
  • This study showed a robust decrease in the need for hospitalization when hydroxychloroquine and azithromycin were prescribed in the early days of symptoms.
  • The best results were observed when treatment was prescribed before day 7 of the initial symptoms, supporting the hypothesis that hydroxychloroquine and azithromycin may act on viral replication.
  • Empirical treatment is performed routinely in several medical illnesses, especially when any delay in the initiation of adequate therapy is potentially harmful.
  • In the study, every 23 empirical treatments performed up to 7 days reduced the need for hospitalization of 1 patient.
  • This study has the limitation of being carried out during a worldwide COVID19 pandemic.
  • It is a therapeutic intervention study, which evaluates the practical use of two drugs (hydroxychloroquine and azithromycin) with a well-known safety profile.


Empirical treatment with hydroxychloroquine associated with azithromycin for suspected cases of COVID-19 infection reduces the need for hospitalization.

Reference Link: https://www.dropbox.com/s/5qm58cd4fneeci2/2020.04.15%20journal%20manuscript%20final.pdf?dl=0

Leave a Reply

Your email address will not be published. Required fields are marked *