Background: Point-of-care ultrasound (POCUS) has become an invaluable tool for quickly diagnosing numerous diseases and procedural guidance. Implementation of POCUS in emergency departments has resulted in reduced length of stay, mortality, and costs. Many applications of POCUS may apply to hospital medicine, yet how well data from other settings applies to inpatient medicine remains unclear.
Methods: We conducted a scoping review of reports published in PubMed Central describing the performance of POCUS among four use cases common in hospital medicine: pneumonia, heart failure, deep vein thromboses (DVTs), and procedural guidance. We report POCUS test characteristics for diagnostic uses as well as impacts on quality metrics where available (e.g., time to hospital discharge). We also discuss the extrapolation from emergency- and acute-care settings to inpatient medicine, highlight existing limitations and knowledge gaps.
Results: The sensitivity of POCUS for pneumonia consistently ranged from 87-94% with specificity ranging 62-82%. Among adults with decompensated heart failure, the sensitivity of POCUS was between 93-100%, with a specificity between 84-96%. Several studies evaluated the utility of serial POCUS evaluations in managing heart failure, and found significant changes in the maximum and minimum inferior vena cava diameter as well as changes in pulmonary B-lines after initiation of diuretic therapy and after hemodialysis. The sensitivity of POCUS for diagnosing DVTs ranged from 90-91% and the specificity ranged from 95-98%, depending on the technique used, including a study which exclusively examined hospitalist performed compression ultrasound. Using POCUS to evaluate for DVTs in emergency departments or intensive care units reduced the time from triage to diagnosis by as much as 14 hours. For abdominal paracenteses, POCUS improved the overall success rate, reduces the risk of attempting the procedure with insufficient abdominal fluid, and reduces complications. POCUS improved success rate of peripheral intravenous catheter insertion by nearly 4-fold. Prior work demonstrated that delays with intravenous access was associated with prolonged time to therapy, diagnostic tests and imaging, and ultimate disposition from the emergency department. The expertise of sonographers varied across studies, from medical trainees (residents and fellows) to experts. Further, few studies described objective competency assessment. How well such findings extrapolate from emergency or intensive care units to the inpatient setting where few standardized POCUS training curricula exist, remains to be determined.
Conclusions: For use cases common to hospital medicine, POCUS demonstrates promising performance characteristics and the potential to improve valuable quality metrics. More work is needed to evaluate the impact of POCUS in the inpatient setting, which would also require system-level efforts to establish POCUS curricula for hospitalists and trainees.