Predicting Fluid Responsiveness by Lung Recruitment Maneuver: One Step Closer to Personalized Perioperative Management

See Article, p 44 The quest for accurate and reliable predictors of intraoperative fluid responsiveness is ongoing in perioperative research. Goal-directed fluid therapy aimed at maintaining intraoperative euvolemia has been shown in multiple settings to minimize postoperative pulmonary and cardiac complications and reduce hospital length of stay.1 During thoracic surgery, a delicate clinical balance must be maintained between restrictive fluid management to prevent pulmonary edema and adequate volume to maintain adequate tissue perfusion. It is essential that we have reliable clinical indices that can determine when intraoperative fluid administration would be beneficial. Given the multitude of variations in patient comorbidities, surgical procedures, and ventilation modalities, finding a universal measure of fluid responsiveness that would apply for all patients is unlikely. Instead, current studies are investigating the predictive value of different dynamic and static indices of fluid responsiveness in specific patient populations, such as those undergoing surgical procedures with one-lung ventilation (OLV). In this issue of Anesthesia & Analgesia, we learn more on predicting fluid responsiveness during OLV in patients ventilated with lung-protective strategies. Kimura et al2 asked whether hemodynamic changes induced by lung recruitment maneuver can predict fluid responsiveness in laterally positioned patients undergoing OLV for thoracoscopic/closed-chest lung resection or esophagectomy. The authors, commendably, studied a cohort of patients undergoing thoracic surgery with OLV in a prospective setting. They monitored the arterial pressure waveform with a commercially available standardized system at a frequency of 100 Hz (100 times/second) and measured mean arterial pressure (MAP), stroke volume (SV), and stroke volume variation (SVV). Pulse pressure variation (PPV) was automatically calculated using their monitor. These static and dynamic hemodynamic indices were then recorded before and at the end of standardized lung recruitment maneuvers and again before and after administration of a volume bolus of 250-mL colloid.2 Kimura et al2 investigated the change in SV (lung recruitment–triggered change in stroke volume [ΔSVRM]) and MAP (lung recruitment–triggered change of mean arterial pressure [ΔMAPRM]) before and after lung recruitment maneuvers to determine fluid responsiveness during OLV by modeling receiver operating characteristic (ROC) curves. Responders were defined as subjects with an increase of ≥10% in MAP or SV on volume administration.2 With robust statistics, Kimura et al2 show for the first time that hemodynamic changes induced by lung recruitment (ΔSVRM and ΔMAPRM) can be used to predict hemodynamic responsiveness to volume boluses during OLV; these changes outperformed the traditionally used SVV and PPV. The Table provides an overview of the hemodynamic indices discussed. Table. - Hemodynamic Indices of Predicting Volume Responsiveness Assessing system Index Technical setup Static MAP Noninvasive: plethysmographic waveform analysis Invasive: arterial line waveform analysis SV Noninvasive: plethysmographic waveform analysis Invasive: arterial line waveform analysis Dynamic PVI Noninvasive: measures plethysmographic waveform variability integrated in some advanced, vendor-specific, hemodynamic monitoring systems Clinical conditions Mechanical ventilation PPV Noninvasive: plethysmographic waveform variability No arrhythmias Invasive: arterial line waveform variability Tidal volume (≥7–8 mL/kg) Computation: (PPmax − PPmin)/PPmean Determined from 3 successive respiratory cycles. Available in some advanced, vendor-specific, hemodynamic systems Heart rate: respiratory rate (HR:RR) ≥ 4 SVV Noninvasive: plethysmographic waveform variability Closed chest Invasive: arterial line waveform variability Computation: (SVmax − SVmin)/SVmean Available in some advanced, vendor-specific, hemodynamic systems Kimura et al2 ΔMAPRM Invasive arterial line waveform Clinical conditions Computation: (MAPPostRM − MAPPreRM/MAPPreRM × 100) OLV Change in MAP before and after lung recruitment maneuver of dependent lung—with an airway pressure of 30 cm H2O for 30 s Tidal volume (6 mL/kg) PEEP (5 cm H2O) ΔSVRM Invasive: arterial line waveform Closed-chest, lateral position Computation: (SVPostRM − SVPreRM/SVPreRM × 100) Change in SV before and after lung recruitment maneuver of dependent lung—with an airway pressure of 30 cm H2O for 30 s. Abbreviations: HR:RR, heart rate to respiratory rate; MAP, mean arterial pressure; MAPPostRM, post-lung recruitment mean arterial pressure; MAPPreRM, pre-lung recruitment mean arterial pressure; OLV, one-lung ventilation; PEEP, positive end expiatory pressure; PPmax, maximum pulse pressure; PPmean, mean pulse pressure; PPmin, minimum pulse pressure; PPV, pulse pressure variation; PVI, plethysmographic variability index; SV, stroke volume; SVmax, maximum stroke volume; SVmean, mean stroke volume; SVmin, minimum stroke volume; SVPostRM, post-lung recruitment stroke volume; SVPreRM, pre-lung recruitment stroke volume; SVV, stroke volume variation; ΔMAPRM, lung recruitment triggered change of mean arterial pressure; ΔSVRM, lung recruitment triggered change in stroke volume. Previous literature addressing dynamic indices in prediction of fluid responsiveness in OLV patients has produced conflicting results.3–6 SVV and PPV are superior to static indices in assessing the adequacy of a patient’s volume status; however, these dynamic indices are significantly affected by heart-lung interactions, ventilation mode, and tidal volume.2 Fu et al3 and Suehiro and Okutani6 both demonstrated the predictive value of SSV and PPV in cohorts of patients undergoing closed-chest thoracoscopic surgery; however, the accuracy was lower in smaller tidal volumes (6 mL/kg) in lung-protective OLV than in traditional ventilation (8 mL/kg). On the other hand, no predictive value of SSV and PPV was found in patients undergoing open-chest thoracotomies or open lobectomies with OLV.4,5 The conflicting results from these studies may be due to variations in patient population, sample size, and surgical conditions; however, the lack of consistent and reproducible results highlights a gap in the current scientific literature regarding how to best predict fluid responsiveness during OLV. The current study by Kimura et al2 begins to address this gap by creatively capitalizing on the hemodynamic changes that occur during standard lung recruitment maneuvers. Protective lung ventilation strategies often include smaller tidal volumes, faster respiratory rates, and intermittent lung recruitment maneuvers to decrease presumed atelectasis. Lung recruitment maneuvers are associated with a reduction in SV, and the magnitude of SV reduction has been shown to be related to the patient’s hypovolemic status and fluid responsiveness.7 Importantly, they show that the magnitude of SV and MAP decrease during a lung recruitment maneuver is a good discriminator of fluid responsiveness, while the traditional indices of SVV and PPV are not.2 The results of this study have the potential for wide clinical applicability. With the advances in minimally invasive thoracic surgeries, such as esophagectomies, lobectomies, and wedge resections, OLV is used more frequently, and therefore, the patient population that the results of this study may apply to is expanding. In addition, potential application of these findings to clinical practice could be quite feasible and may not require additional equipment or new clinical procedures, as lung recruitment maneuvers and invasive blood pressure monitoring are routinely performed in these cases.8 In addition, high-risk surgical patients have been shown to benefit from careful optimization of fluid administration,9 and it has recently been proposed that goal-directed fluid strategies be applied to patients undergoing thoracic surgery specifically to maintain optimal organ perfusion.10 However, there are also limitations to the study that should be kept in mind. To maintain a comparable patient cohort and minimize potentially confounding factors, Kimura et al2 excluded subjects suffering from systolic (ejection fraction ≤55%) or diastolic (early/late diastolic transmitral flow velocity <0.8 and early diastolic mitral annular velocity <8 cm/s) cardiac dysfunction or right heart failure. To prove the value of the current findings, future studies will have to assess the results in patients suffering from cardiac dysfunction. Additionally, dynamic fluid indices have been shown to be affected by cyclic heart-lung interactions and open-chest conditions, in which intrathoracic pressure changes associated with mechanical ventilation are reduced due to exposure of the chest cavity to normal ambient pressure. Kimura et al2’s indices of ΔMAPRM and ΔSVRM were not evaluated in open-chest conditions. As previously mentioned, the cumulative studies performed in open-chest settings have shown no predictive value for PPV and SSV. Many thoracic surgical procedures are either performed open or have the possibility of converting into open if the thoracoscopic approach is not successful. Further research is needed to find indices to predict fluid responsiveness during open-chest surgery. How the results of this study would apply to an open-chest model or different kinds of body positions11 is unknown. Furthermore, the utility of recruitment maneuver-induced indices of fluid responsiveness with double-lung ventilation in either closed-chest or open abdominal cases will also need to be determined to truly understand the broader translatability of these findings. This prospective study in a standardized cohort provides a better understanding of cardiopulmonary interactions during OLV and reveals how currently used anesthetic maneuvers can be further utilized to improve patient care. Personalized perioperative care involves more than caring for individual organ systems; rather, it benefits from a passion for managing their complex interplay. As Kehlet and Wilmore12 defined it a decade ago, several small improvements without significant effect in isolation may sum up synergistically and improve the outcome. The article by Kimura et al2 inspires us to do so, 1 step at a time. DISCLOSURES Name: Kimberly Howard-Quijano, MD, MS, FASE. Contribution: This author helped write this editorial and see, review, and approve the final manuscript. Name: Markus M. Luedi, MD, MBA. Contribution: This author helped write this editorial and see, review, and approve the final manuscript. This manuscript was handled by: Nikolaos J. Skubas, MD, DSc, FACC, FASE.