Statistical biases due to anonymization evaluated in an open clinical dataset from COVID-19 patients.

Anonymization has the potential to foster the sharing of medical data. State-of-the-art methods use mathematical models to modify data to reduce privacy risks. However, the degree of protection must be balanced against the impact on statistical properties. We studied an extreme case of this trade-off: the statistical validity of an open medical dataset based on the German National Pandemic Cohort Network (NAPKON), which was prepared for publication using a strong anonymization procedure. Descriptive statistics and results of regression analyses were compared before and after anonymization of multiple variants of the original dataset. Despite significant differences in value distributions, the statistical bias was found to be small in all cases. In the regression analyses, the median absolute deviations of the estimated adjusted odds ratios for different sample sizes ranged from 0.01 [minimum = 0, maximum = 0.58] to 0.52 [minimum = 0.25, maximum = 0.91]. Disproportionate impact on the statistical properties of data is a common argument against the use of anonymization. Our analysis demonstrates that anonymization can actually preserve validity of statistical results in relatively low-dimensional data.

Safety and preliminary efficacy of sequential multiple ascending doses of solnatide to treat pulmonary permeability edema in patients with moderate to severe ARDS in a randomized, placebo-controlled, double-blind trial: preliminary evaluation of safety and feasibility in light of the COVID-19 pandemic.

BACKGROUND: In May 2018, the first patient was enrolled in the phase-IIb clinical trial "Safety and Preliminary Efficacy of Sequential Multiple Ascending Doses of Solnatide to Treat Pulmonary Permeability Edema in Patients with Moderate to Severe ARDS." With the onset of the COVID-19 pandemic in early 2020, the continuation and successful execution of this clinical study was in danger. Therefore, before the Data Safety Monitoring Board (DSMB) allowed proceeding with the study and enrollment of further COVID-19 ARDS patients into it, additional assessment on possible study bias was considered mandatory. METHODS: We conducted an ad hoc interim analysis of 16 patients (5 COVID-19- ARDS patients and 11 with ARDS from different causes) from the phase-IIB clinical trial. We assessed possible differences in clinical characteristics of the ARDS patients and the impact of the pandemic on study execution. RESULTS: COVID-19 patients seemed to be less sick at baseline, which also showed in higher survival rates over the 28-day observation period. Trial specific outcomes regarding pulmonary edema and ventilation parameters did not differ between the groups, nor did more general indicators of (pulmonary) sepsis like oxygenation ratio and required noradrenaline doses. CONCLUSION: The DSMB and the investigators did not find any evidence that patients suffering from ARDS due to SARS-CoV-2 may be at higher (or generally altered) risk when included in the trial, nor were there indications that those patients might influence the integrity of the study data altogether. For this reason, a continuation of the phase IIB clinical study activities can be justified. Researchers continuing clinical trials during the pandemic should always be aware that the exceptional circumstances may alter study results and therefore adaptations of the study design might be necessary.

[Characteristics and outcome of 70 ventilated COVID-19 patients : Summary after the first wave at a university center]. / Charakteristika und Outcome von 70 beatmeten COVID-19-Patienten : Bilanz nach der ersten Welle an einem universitären Zentrum.

BACKGROUND: In a pandemic situation the overall mortality rate is of considerable interest; however, these data must always be seen in relation to the given healthcare system and the availability of local level of care. A recently published German data evaluation of more than 10,000 COVID-19 patients treated in 920 hospitals showed a high mortality rate of 22% in hospitalized patients and of more than 50% in patients requiring invasive ventilation. Because of the high infection rates in Bavaria, a large number of COVID-19 patients with considerable severity of disease were treated at the intensive care units of the LMU hospital. The LMU hospital is a university hospital and a specialized referral center for the treatment of patients with acute respiratory distress syndrome (ARDS). OBJECTIVE: Data of LMU intensive care unit (ICU) patients were systematically evaluated and compared with the recently published German data. METHODS: Data of all COVID-19 patients with invasive and noninvasive ventilation and with completed admission at the ICU of the LMU hospital until 31 July 2020 were collected. Data were processed using descriptive statistics. RESULTS: In total 70 critically ill patients were included in the data evaluation. The median SAPS II on admission to the ICU was 62 points. The median age was 66 years and 81% of the patients were male. More than 90% were diagnosed with ARDS and received invasive ventilation. Treatment with extracorporeal membrane oxygenation (ECMO) was necessary in 10% of the patients. The median duration of ventilation was 16 days, whereby 34.3% of patients required a tracheostomy. Of the patients 27.1% were transferred to the LMU hospital from external hospitals with reference to our ARDS/ECMO program. Patients from external hospitals had ARDS of higher severity than the total study population. In total, nine different substances were used for virus-specific treatment of COVID-19. The most frequently used substances were hydroxychloroquine and azithromycin. Immunomodulatory treatment, such as Cytosorb® (18.6%) and methylprednisolone (25.7%) were also frequently used. The overall in-hospital mortality rate of ICU patients requiring ventilation was 28.6%. The mortality rates of patients from external hospitals, patients with renal replacement therapy and patients with ECMO therapy were 47.4%, 56.7% and 85.7%, respectively. CONCLUSION: The mortality rate in the ventilated COVID-19 intensive care patients was considerably different from the general rate in Germany. The data showed that treatment in an ARDS referral center could result in a lower mortality rate. Low-dose administration of steroids may be another factor to improve patient outcome in a preselected patient population. In the authors' opinion, critically ill COVID-19 patients should be treated in an ARDS center provided that sufficient resources are available.

Safety and preliminary efficacy of sequential multiple ascending doses of solnatide to treat pulmonary permeability edema in patients with moderate-to-severe ARDS-a randomized, placebo-controlled, double-blind trial.

BACKGROUND: Acute respiratory distress syndrome (ARDS) is a complex clinical diagnosis with various possible etiologies. One common feature, however, is pulmonary permeability edema, which leads to an increased alveolar diffusion pathway and, subsequently, impaired oxygenation and decarboxylation. A novel inhaled peptide agent (AP301, solnatide) was shown to markedly reduce pulmonary edema in animal models of ARDS and to be safe to administer to healthy humans in a Phase I clinical trial. Here, we present the protocol for a Phase IIB clinical trial investigating the safety and possible future efficacy endpoints in ARDS patients. METHODS: This is a randomized, placebo-controlled, double-blind intervention study. Patients with moderate to severe ARDS in need of mechanical ventilation will be randomized to parallel groups receiving escalating doses of solnatide or placebo, respectively. Before advancing to a higher dose, a data safety monitoring board will investigate the data from previous patients for any indication of patient safety violations. The intervention (application of the investigational drug) takes places twice daily over the course of 7 days, ensued by a follow-up period of another 21 days. DISCUSSION: The patients to be included in this trial will be severely sick and in need of mechanical ventilation. The amount of data to be collected upon screening and during the course of the intervention phase is substantial and the potential timeframe for inclusion of any given patient is short. However, when prepared properly, adherence to this protocol will make for the acquisition of reliable data. Particular diligence needs to be exercised with respect to informed consent, because eligible patients will most likely be comatose and/or deeply sedated at the time of inclusion. TRIAL REGISTRATION: This trial was prospectively registered with the EU Clinical trials register (clinicaltrialsregister.eu). EudraCT Number: 2017-003855-47 .