A Novel Lung Ultrasound Training Program to Predict Severity
of Acute Lower Respiratory Tract Infections in Sri Lanka
Trainees without prior lung ultrasound
experience can be taught to perform
quality scans with limited time and
Sky Vanderburg1, Iroshani Kodikara2, Muhunthan Sellathurai3,
Champika Bodinayake2,3, Ajith Nagahawatte2, Larry Park4,5,
Christopher Woods4,5, Elisabeth Riviello6, L. Gayani Tillekeratne4,5
1. Hubert-Yeargan Center for Global Health, Duke University, USA.
2. Faculty of Medicine, University of Ruhuna, Sri Lanka.
3. Teaching Hospital Karapitiya, Sri Lanka.
4. Duke Global Health Institute, USA.
5. Department of Medicine, Duke University, USA
6. Beth Israel Deaconess Medical Center and Harvard Medical School, USA.
Introduction and Objectives
•Low- and middle-income countries (LMICs) bear a disproportionate
burden of lower respiratory tract infections (LRTIs)
•Consequently, timely diagnosis and allocation of scarce resources is
crucial in LRTI cases to reduce morbidity and mortality.
•Bedside LUS has demonstrated promise for improving LRTI care but
successful training methods in LMICs are not yet established.
•A LUS training program and pilot study was conducted to describe the
feasibility characteristics and inter-operator agreement of LUS
Each trainee received approximately 24 hours of training and performed 40-45
LUS examinations, of which 3-6 were audited. After program completion, the
average LUS examination duration was 10 minutes. The average and
individual intraclass correlation coefficient (ICC) was 0.95 (0.87, 0.98) and
0.91 (0.78, 0.96), respectively, with a trend toward improved trainee
performance with each successive audit.
Five clinical trainees without prior LUS experience underwent training
sessions conducted by a LUS expert which included instruction in LUS
basics, a 12-zone protocol, and LRTI severity score determination. Image
acquisition and interpretation were taught using open source materials
along with practical training on volunteer patients. Trainees then performed
LUS examinations on inpatients while blinded to radiographic and other
clinical data. The expert then audited trainees by repeating a subset of
examinations for comparison.
Quality LUS training in Sri Lanka is feasible with minimal resources and
has promising implications to improve diagnostic and predictive capacity
in cases of LRTI.
Figure 2. Trainee and Expert Scores Converge within +/- 2 Points by 20th Exam
Figure 3. Scores of Trainees and Expert Demonstrate Excellent Correlation
Indication for LUS
o Supplemental oxygen
o Nasal cannula
o Non–rebreathing or Venturi mask
o Mechanical ventilation with a
Oxygenation parameters (closest to
time of performing LUS)
: % or L/min
Cardiac failure or fluid overload
Clinical insult in the last week
o Infection unknown
o Surgical procedure
o Massive transfusion
o Drug abuse/overdose
o Inhalation trauma
o Myocardial infarction
o Other, please specify:
Ventilator settings (at moment of performing LUS)
o Volume–controlled ventilation
o Pressure–controlled ventilation
o Pressure support
o Adaptive support ventilation
o Non–invasive ventilation
Ventilator settings and parameters
Pmax: cm H
PEEP: cm H
RR: breaths/min (set)
RR: breath/min (measured)
Lung ultrasound report form:
A, A–pattern (only A lines visible); B, B–pattern (> 2 B–lines or confluent B–lines); C,
C–pattern (consolidated lung tissue); E–pattern (pleural effusion); 0–pattern (absence
of lung sliding) – C and E can be present at same time
Patient name DOB / / Patient ID
Examiner Date / / Time : HH
R3 R1 L1 L3
R5 R4 R2 R2 L4 L5
Figure 1. 12 Zones for Lung Ultrasound
Research Funded By:
•Thrasher Research Fund Early Career Award
•NIH Fogarty Global Health Fellowship