Background: Surgical patients represent a highly vulnerable patient group, who are at particular risk of COVID-19 exposure and complications whilst in hospital for essential surgical treatment. They are vulnerable because of their underlying comorbidity and also because they will be subjected to artificial ventilation at the time of surgery. There are currently no interventional trials looking to prevent or mitigate the pulmonary complications associated with concurrent COVID-19 infection acquired either just before surgery or during the postoperative stay in hospital. Primary objectives: To provide reliable estimates of the effect of study treatments on postoperative pulmonary complications during the COVID-19 pandemic. Secondary objectives: To assess the effects of study treatments on: · Post-operative proven COVID-19 pulmonary complications · Overall SARS-CoV-2 infected rate · Duration of intensive care and total hospital stay · Pulmonary function in keeping with WHO Solidarity Trial outcome scale (detailed in Appendix 2) · Safety and tolerability of study treatments Design: Adaptive platform design, multi-centre, open-label, randomised controlled trial. The interim trial results will be monitored by an independent Data Monitoring Committee (DMC), who will periodically assess whether the randomised comparisons in the study have provided evidence on postoperative pulmonary complications that is strong enough to influence global treatment guidelines. Trial arms will be amended on recommendation from the DMC and new arms can be added if a new drug or vaccine is released during the study that requires evaluation. Centre eligibility: Any hospital performing elective or emergency adult surgery that has recorded at least one case of COVID-19. Participants Inclusion: Adults aged 16 years and over listed to undergo any type of inpatient surgery requiring general or regional anaesthesia (such as vulnerable patients undergoing surgery for a fractured neck of femur). who are asymptomatic of COVID-19 infection (including patients with: those not tested, negative test results, positive test but no symptoms) and are able to give informed consent. Exclusion: Procedures under local anaesthesia; patients who have symptoms of COVID-19 infection (by confirmed COVID-19 test or a clinical diagnosis); existing regular preoperative treatment with trial drugs; known history of adverse reaction or contraindication to trial drugs; pregnant patients (including caesarean section). Interventions and randomisation: The trial drugs are Lopinavir-Ritonavir and Hydroxychloroquine. Patients will be randomised 1:1:1:1 to (A) Control (normal practice; neither trial drug), (B) Lopinavir-Ritonavir only, (C) Hydroxychloroquine only, (D) both Lopinavir-Ritonavir and Hydroxychloroquine. Primary Outcome: Any one of the following COVID-19 specific, inpatient, postoperative pulmonary complications: pneumonia; acute respiratory distress syndrome (ARDS); or death. Secondary Outcomes: · Unexpected ventilation (unexpected inability to extubate and wean patient from ventilation after general anaesthesia, or reintubation and ventilation by 30 days after surgery) · Postoperative diagnosis of proven COVID-19 pulmonary complications · Overall SARS-CoV-2 infected rate (symptomatic and/or asymptomatic) · Duration of hospital stay (including time spent in intensive care, time ventilated) · Pulmonary function in keeping with the WHO Solidarity Trial outcome scale Sample size: The trial uses a Bayesian adaptive platform design, allowing for termination of arms if their superiority to standard care is established, and addition of new treatment arms. The detailed statistical design of the trial will be described in a Statistical Analysis Plan. The adaptive platform design does not have a fixed sample size. We will set a maximum sample size for each arm of 1600; which is similar to the number needed for a traditional frequentist 2-arm comparison to achieve p<0.05 with 90% power, with event rates of 16% in the control group and 12% in the intervention arm. Interim analyses will be conducted every 200 patients, starting when 250 patients per arm have reached the end of follow-up. All analyses will use Bayesian methods, with weakly informative priors, which will allocate very low probability to unrealistic treatment effects. New treatment arms will be added as more interventions are proposed and they will be compared only with controls recruited contemporaneously. Data from earlier control patients will be used in a more informative prior for the control group in these comparisons. |