A Comparison of Micropore Membrane Inlet Mass Spectrometry–Derived Pulmonary Shunt Measurement with Riley Shunt in a Porcine Model Bastian Duenges, PhD* Andreas Vogt, MD† Marc Bodenstein, MD* Hemei Wang, MD, PhD*‡ Stefan Bo ¨hme, MS* Bernd Ro ¨ hrig, PhD§ James E. Baumgardner, MD, PhD Klaus Markstaller, MD* BACKGROUND: The multiple inert gas elimination technique was developed to measure shunt and the ratio of alveolar ventilation to simultaneous alveolar capillary blood flow in any part of the lung (V A '/Q') distributions. Micropore membrane inlet mass spectrometry (MMIMS), instead of gas chromatography, has been introduced for inert gas measurement and shunt determination in a rabbit lung model. However, agreement with a frequently used and accepted method for quantifying deficits in arterial oxygenation has not been established. We compared MMIMS-derived shunt (M-S) as a fraction of total cardiac output (CO) with Riley shunt (R-S) derived from the R-S formula in a porcine lung injury model. METHODS: To allow a broad variance of atelectasis and therefore shunt fraction, 8 sham animals did not receive lavage, and 8 animals were treated by lung lavages with 30 mL/kg warmed lactated Ringer’s solution as follows: 2 animals were lavaged once, 5 animals twice, and 1 animal 3 times. Variables were recorded at baseline and twice after induction of lung injury (T1 and T2). Retention data of sulfur hexafluoride, krypton, desflurane, enflurane, diethyl ether, and acetone were analyzed by MMIMS, and M-S was derived using a known algorithm for the multiple inert gas elimination technique. Standard formulas were used for the calculation of R-S. RESULTS: Forty-four pairs of M-S and R-S were recorded. M-S ranged from 0.1% to 35.4% and R-S from 3.7% to 62.1%. M-S showed a correlation with R-S described by linear regression: M-S =-4.26 + 0.59  R-S (r 2 = 0.83). M-S was on average lower than R-S (mean =-15.0% CO, sd = 6.5% CO, and median =-15.1), with lower and upper limits of agreement of -28.0% and -2.0%, respectively. The lower and upper limits of the 95% confidence intervals were -17.0 and -13.1 (P  0.001, Student’s t-test). CONCLUSIONS: Shunt derived from MMIMS inert gas retention data correlated well with R-S during breathing of oxygen. Shunt as derived by MMIMS was generally less than R-S. (Anesth Analg 2009;109:1831–5) The multiple inert gas elimination technique (MIGET) is an established and well-accepted method for determining pulmonary shunt and the ratio of alveolar ventilation to simultaneous alveolar capillary blood flow in any part of the lung (V A '/Q') distribu- tions. 1–13 MIGET has provided useful insights into the mechanisms of gas exchange pathophysiology in many experimental studies. 1–7,9 –15 A drawback of the method, however, is a time-consuming analysis of the inert gases by gas chromatography. The recently in- troduced micropore membrane inlet mass spectrom- etry (MMIMS) measures inert gas partial pressures rapidly in small blood samples with little chance for technical errors. 14 MMIMS-derived shunt (M-S) frac- tion, however, has not been validated against a refer- ence method for shunt measurement. Calculation of Riley shunt (R-S) from arterial, mixed venous, and assumed pulmonary end-capillary blood gas data is a frequently used method for quan- tifying deficits in arterial oxygenation. 4,9,12,16 R-S com- bines the contributions of shunt (0 V A '/Q') with the contributions of low V A '/Q' lung units to explain hypoxemia in terms of a single hypothetical admixture compartment and gives little information about the From the *Department of Anesthesiology, Johannes Gutenberg- University, Mainz, Germany; †University Department of Anesthesiology and Pain Therapy, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; ‡Beijing Institute of Pharmacology and Toxi- cology, Beijing, China; §Institute of Medical Biostatistics, Epidemiology and Informatics (IMBEI), Johannes Gutenberg-University, Mainz, Ger- many; and Oscillogy, LLC, Folsom, Pennsylvania. Accepted for publication July 28, 2009. Supported by German Research Foundation DFG Ma 2398/3, Swiss National Foundation SNF POIB—117065/1 and an institu- tional grant of the Department of Anesthesiology, Inselspital, Bern University Hospital, and University of Bern. J. E. Baumgardner is President of Oscillogy LLC, the manufac- turer of the MIGET by MMIMS system. Address correspondence and reprint requests to Andreas Vogt, MD, University Department of Anesthesiology and Pain Therapy, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland. Address e-mail to andreas.vogt@insel.ch. Copyright © 2009 International Anesthesia Research Society DOI: 10.1213/ANE.0b013e3181bbc401 Vol. 109, No. 6, December 2009 1831