Estimation of intracardiac shunts in young children with a novel indicator dilution technology

Authors, Journal, Affiliations, Type, DOI

• Authors: Theodor Skuli Sigurdsson, Lars Lindberg
• Journal: Scientific Reports (Nature Publishing Group), 2020;10:1337
• Affiliations: Department of Pediatric Anesthesia and Intensive Care, Children's Hospital, University Hospital of Lund, Sweden; Department of Anesthesia and Intensive Care, Landspítalinn, University Hospital of Iceland
• Type: Single-center, prospective, observational method-comparison study
• DOI: https://doi.org/10.1038/s41598-020-58347-2

Overview

A prospective single-center study (n=44, mean age 12 months, weight 2.7–13.6 kg) validated the COstatus monitor — a minimally invasive transpulmonary ultrasound dilution device — for detecting and quantifying intracardiac shunts (Qp/Qs ratio) before and after corrective ASD/VSD surgery. COstatus demonstrated excellent diagnostic accuracy (sensitivity 95.7%, specificity 97.6%, AUC 0.97) for detecting left-to-right shunts. However, COstatus significantly underestimated Qp/Qs ratios in moderate and small shunt groups compared to perivascular flow probes and the oximetric shunt equation, likely due to algorithm limitations in the young pediatric age group. The device requires only existing arterial and central venous lines and adds hemodynamic parameters (cardiac output, central blood volumes) beyond what echocardiography alone provides.

Keywords

Intracardiac shunt, Qp/Qs ratio, indicator dilution, COstatus, transpulmonary ultrasound dilution, atrial septal defect, ventricular septal defect, pediatric hemodynamics, cardiac output monitoring

Key Takeaways

Background

• Clinical evaluation of hemodynamics in critically ill children is often imprecise; reliable monitoring devices are needed
• Undiagnosed intracardiac shunts are associated with significant morbidity
• Echocardiography detects the presence of intracardiac shunts but cannot accurately estimate the Qp/Qs ratio
• MRI can estimate Qp/Qs but is impractical in critically ill children
• Standard Qp/Qs methods (cardiac catheterization, oximetric shunt equation, perivascular flow probes) are highly invasive and require general anesthesia

Qp/Qs Ratio — Definition and Clinical Significance

• <1.0 = Right-to-left shunt
• 1.0 = No shunt
• 1.0–1.5 = Small left-to-right shunt
• 1.5–2.0 = Moderate left-to-right shunt
• ≥2.0 = Large left-to-right shunt; generally considered indication for surgical correction
• Qp/Qs determination is used to assess clinical significance and timing of surgical correction

COstatus Monitor — Mechanism

• Uses an extracorporeal AV loop connected to existing arterial (radial) and central venous (jugular) catheters; no additional vascular access required
• A roller-pump maintains constant blood flow through the AV loop at 9–12 mL/min
• Isotonic saline (body temperature, 0.5–1.0 mL/kg) injected into the venous port creates a dilution bolus
• Blood and saline have different ultrasound velocities (blood 1,560–1,585 m/s; saline 1,530 m/s) — saline dilution causes a detectable fall in ultrasound velocity
• Arterial sensor detects the transcardiopulmonary dilution curve; asymmetry in the descending limb (delayed descent) indicates indicator recirculating through the pulmonary circuit due to L→R shunt
• Normal a:b ratio (ascending limb distance from peak / descending limb distance from peak, measured at half-peak level) ≈ 1.4 ± 13%
• Qp/Qs formula: ≈ (b/a)/1.4 ± 13%; displayed as categorical range (five groups per Table 1, not a continuous value)
• Additionally measures cardiac output, central blood volume, and active circulation blood volume

Results — Shunt Detection

• COstatus detected Qp/Qs >1.0 in 43/44 patients with TEE-confirmed defects pre-operatively
• Sensitivity 95.7% (95% CI 85.5–99.5%); specificity 97.6% (95% CI 87.1–99.9%); AUC 0.97
• 2/44 patients had bidirectional shunts detected by COstatus — excluded from L→R comparison
• Post-operatively: 3 very small residual defects detected by TEE color Doppler; COstatus identified 2/3 + 1 additional defect not seen by TEE (corroborated by oximetric calculation)
• Residual shunt incidence 7% post-repair; all considered clinically insignificant

Results — Shunt Ratio Estimation

• Mean Qp/Qs PVFP: 2.9 ± 1.2 (range 1.0–9.7)
• Mean Qp/Qs OSE: 3.0 ± 1.6 (range 1.1–10.0)
• PVFP vs OSE bias: 0.07 (Bland-Altman); percentage error 58.8% — high variability even between reference methods
• COstatus significantly underestimated Qp/Qs in moderate and small shunt groups vs both PVFP and OSE (p<0.05)
• In large shunt group (Qp/Qs >2), COstatus reports only a categorical ">2" — direct numerical comparison with reference methods impossible

Comparison with Prior Dilution Studies

• Historical dilution methods in children overestimated Qp/Qs vs oximetric techniques — caused by rapid systemic recirculation inflating the area under the curve
• COstatus underestimates Qp/Qs — opposite direction of error; reflects corrected AUC algorithm that avoids overestimation but overshoots in the other direction
• In adults, oximetric technique produces higher Qp/Qs estimates than dilution — consistent with COstatus findings in this pediatric study

Methods Detail

• 44 children undergoing elective cardiac surgery for ASD and/or VSD; inclusion: parental consent + weight <15 kg + ASD/VSD defect; exclusion: perioperative arrhythmias, valvular regurgitation
• Five consecutive COstatus measurements paired with five PVFP measurements (sequential, not simultaneous); COstatus shunt positive if indicated in ≥2/5 measurements
• Reference 1 — PVFP: transit-time ultrasound probes around ascending aorta (Qs) and pulmonary truncus (Qp); placed consecutively (simultaneous placement compromised circulation)
• Reference 2 — OSE: blood gases from pulmonary truncus, IVC, SVC, and arterial catheter; Flamm's formula for mixed venous saturation: (3×SVC + IVC)/4; FiO₂ 0.3–0.4 to ensure full arterial saturation
• TEE performed by cardiologist blinded to other findings for pre- and post-operative shunt assessment
• Statistical: ANOVA for between-method comparison; Bland-Altman for PVFP vs OSE bias; ROC for shunt detection accuracy

Limitations of the Document

• Single-center study; small sample (n=44)
• Flow probes placed sequentially rather than simultaneously — introduces temporal variability in hemodynamic state between Qp and Qs measurements
• Aortic flow probe excludes coronary blood flow (~7% of cardiac output; higher in hearts with L→R shunt) → underestimates Qs → inflates reference Qp/Qs
• Oximetric technique relies on steady-state oxygen uptake = consumption assumption; Flamm's formula may produce false-low mixed venous saturation values
• COstatus presents Qp/Qs as a categorical range value — prevents precise quantitative validation
• Algorithm likely requires age-specific optimization for young children (heart rate, ventricular function, and peripheral resistance differ substantially vs adults)
• Limited to simple ASD/VSD; results may not generalise to complex mixing lesions (single ventricle physiology)
• No continuous arterial pressure monitoring during COstatus measurement sessions

Key Concepts Mentioned

• concepts/Intracardiac-Shunts — primary subject; Qp/Qs ratio as index of shunt magnitude and surgical indication
• concepts/Cardiac-Output-Measurement — Stewart-Hamilton principle; COstatus as ultrasound dilution variant
• concepts/Atrial-Septal-Defect — most common single defect type in cohort; Qp/Qs-guided closure decision context

Key Entities Mentioned

• entities/COstatus-Monitor — Transonic Systems Inc., Ithaca, NY; novel ultrasound dilution monitor; AV loop design

Wiki Pages Updated

• wiki/sources/shunt-nature-sr-2020.md — created (this page)
• wiki/concepts/Intracardiac-Shunts.md — created
• wiki/concepts/Indicator-Dilution-Cardiac-Output.md — updated (added transpulmonary ultrasound dilution/COstatus section)
• wiki/concepts/Atrial-Septal-Defect.md — updated (added Qp/Qs estimation context under Diagnostic Approach)
• wiki/sourceindex.md — updated
• wiki/wikiindex.md — updated