The proposed study is a prospective cohort study in which a select group of emergency physicians at Komfo Anokye Teaching Hospital (KATH) in Ghana will be trained in cardiopulmonary ultrasound (CPUS). Following the training, patients who present to the ED with undifferentiated shock and/or dyspnea will either receive usual and customary care supplemented with cardiopulmonary ultrasonography-guided diagnosis and treatment during their initial resuscitation, or usual and customary care alone depending on whether the treating physician has received CPUS training. The main outcomes is the impact of CPUS on correct diagnosis. Information regarding initial treatment strategies, diagnoses and 24-hour mortality will be collected via manual review of paper charts and medical records.

Methods:

The training protocol and pilot study have been developed through a collaborative effort involving emergency medicine physicians at University of Michigan (UM) and KATH. Select resident physicians (approximately 1/3 of the emergency medicine resident physicians) at KATH will participate in a one-week, hands-on ultrasound course concentrating on solidifying their preexisting skills in ultrasound, with a focus on the RUSH and BLUE protocols for assessment of critically ill patients. Resident physicians will be trained, as attending physicians are not routinely present during the majority of patient evaluations.

The RUSH (Rapid Ultrasound in SHock) protocol was first described in publication in 2010 and has since become widely used throughout the country to aid in identifying the cause of shock for a patient. It is a three-step ultrasound scanning protocol, first evaluating "the pump" or heart, second "the tank" or intravascular volume status, and third "the pipes" or large arteries and veins. Ultrasound findings from the exam are then compared with those typically seen in hypovolemic shock, cardiogenic shock, obstructive shock and distributive shock (1). The BLUE (Bedside Lung Ultrasound in Emergency) protocol was initially published in 2008 as a quick tool to determine diagnosis in acute respiratory failure. Based on ultrasound findings, this scanning protocol provides specific profiles for six main causes of acute respiratory distress, pulmonary edema, pulmonary embolism, pneumonia, chronic obstructive pulmonary disease, asthma, and pneumothorax (2).

As most of the emergency physicians at KATH have previously expressed interest in receiving this training, we will offer training to all physicians in three different cohorts, 1/3 initially, 1/3 part way through the study, and 1/3 at the end. The physicians in the initial training group will be chosen based off their availability and work schedule. The participants will take a test prior to beginning the training and also at the end. They will also be required to pass an OSCE (Objective Structured Clinical Exam) at the end of the week. One week will be allocated for ultrasound training, to allow the physicians to participate and practice their new skills. These select residents will make up the cohort of CPUS-trained physicians.

The pilot study will be conducted over a four-month period in the emergency department at KATH. All patients presenting to the ED will be screened for study inclusion criteria as defined below by a triage nurse, and informed consent will be obtained by research assistants in all patients deemed competent.

All patients will immediately receive usual and customary care, to include physical examination, IV access, continuous cardiac monitoring, and supplemental oxygen. Following the initial assessment, the treating physician will be asked to choose the most likely diagnosis from a provided checklist. All checklists were designed specifically for research purposes in this study and the information documented on the checklists will be used in clinical decision-making. The information contained in the checklists is based on the well-validated RUSH and BLUE protocols and will be used strictly in the context of these protocols only. The physicians filling out the checklists will be part of the study team at KATH. When a patient meets inclusion criteria and their treating physician has received CPUS-training, the patient will receive a CPUS exam in addition to standard care. In addition to the CPUS exam, any other ultrasound exams deemed necessary for clinical care will occur as they normally do as part of regular standard of care. Ultrasound will not be withheld for any patient as part of their clinical care. There will be no aspects of the study conducted prior to enrollment that will impact care or study design.

Patients in the CPUS group will receive imaging in concordance with the standardized RUSH and BLUE protocols and findings will be documented on a second checklist. Emergency physicians trained in CPUS will perform the ultrasound exams, after which they will again choose the most likely diagnosis from the same initial checklist. The pre-ultrasound and post-ultrasound diagnoses will be recorded for comparison. Diagnostic accuracy will be assessed based on post-hoc review of hospital records. Ultrasound competency will be assessed intermittently by means of direct supervision during site visits to ensure that there is no skill degradation and to ensure safety of study participants.

Data Analysis:

Power analysis was performed for the primary outcome of diagnostic accuracy. Assuming a baseline of 60% correct diagnosis we calculated that 158 subjects are required to detect an absolute 30% increase in correct diagnosis. A minimum of 180 patients will be enrolled to account for loss to follow-up. Additionally, the information gathered during the pilot study will be used to help guide planning for a long-term follow-up study in the future. During the initial ultrasound training, competency will be assessed using pre and post-tests, in addition to OSCE (Objective Structured Clinical Exam). General Linear Model (GLM) framework, with pretest scores as the covariate, will be used to analyze the pre and post-test data. All data from the pilot study will be collected on I-pads or paper copies and transferred into a CSV-formatted file by a research assistant. Data will be examined using SPSS statistical analysis software using linear and (multiple) logistic regression analysis. A p-value of <0.05 will be interpreted as statistically significant.

Ethical Considerations:

Ethics approval will be sought from the Committee for Human Research, Publication and Ethics (CHRPE) of the School of Medical Sciences, Kwame Nkrumah University of science and Technology (S.M.S.-K.N.U.S.T.) and the Institutional Review Board (IRB) of the University of Michigan.

All patients considering taking part in this study will be provided with a consent form describing the interventions and given sufficient time to decide whether or not to participate. When informed consent cannot be obtained from the patient due to incapacitation or impaired decision-making, consent will be obtained from the next of kin or from legally authorized representative. Legal guardians or the head of household can serve as a proxy or legally authorized representative. According to local customs, in the case where there are no parents then the "head of the family" is given proxy. This person may not have court documentation but is presented by the family and accepted as the one to make decisions. Only adults, those over the age of 18, will be included in this study. The subject population consists of the community surrounding KATH which sees a range of patients, from children to adults. Only adults meeting the inclusion criteria will be included in this study.

Confidentiality will be maintained by de-identifying patient information as early as possible during the study and storing personal information and data on secure servers in the United States which will likely be very effective in protecting patient information. In reports, data and discussion are confidential.

Inclusion Criteria:

Presence of at least one of the following signs or symptoms of hypoperfusion or hypoxia:

• Unresponsiveness or altered mental status with a GCS <13
• Diaphoresis
• Capillary refill >3 seconds
• Systolic blood pressure <100 at any point between arrival to the ED and IV fluid administration
• Tachycardia >100 bpm
• Tachypnea >20/min
• Pulse oximetry of <92% at any point between arrival to the ED and administration of supplemental oxygen

Exclusion Criteria:

• Chronic low blood pressure, as evidenced by patient report or documentation
• ACS, determined by ST elevation on EKG (troponin levels not easily available at study site)
• Significant resuscitative measures prior to enrollment, including defibrillation, ALS medications, or mechanical ventilation
• Determination of etiology of shock prior to enrollment, such as obvious signs of major trauma or obvious GI bleeding
• Onset of signs or symptoms of shock after initial evaluation by a physician