Near-infrared spectroscopy for kidney oxygen monitoring in a porcine model of hemorrhagic shock, hemodilution, and REBOA.

Acute kidney injury is a common complication of trauma and hemorrhagic shock. In a porcine model of hemorrhagic shock, resuscitative endovascular balloon aortic occlusion (REBOA) and hemodilution, we hypothesized that invasive kidney oxygen concentration measurements would correlate more strongly with noninvasive near infra-red spectroscopy (NIRS) oxygen saturation measurements when cutaneous sensors were placed over the kidney under ultrasound guidance compared to placement over the thigh muscle and subcutaneous tissue. Eight anesthetized swine underwent hemorrhagic shock 4 of which were resuscitated with intravenous fluids prior to the return of shed blood (Hemodilution protocol) and 4 of which underwent REBOA prior to resuscitation and return of shed blood (REBOA protocol). There was a moderate correlation between the NIRS and kidney tissue oxygen measurements (r = 0.61 p < 0.001; r = 0.67 p < 0.001; r = 0.66 p < 0.001for left kidney, right kidney, and thigh NIRS respectively). When the animals were separated by protocol, the Hemodilution group showed a weak or nonsignificant correlation between NIRS and kidney tissue oxygen measurements (r = 0.10 p < 0.001; r = 0.01 p = 0.1007; r = 0.28 p < 0.001 for left kidney, right kidney, and thigh NIRS respectively). This contrasts with the REBOA group, where left and right kidney as well as thigh NIRS were moderately correlated with kidney tissue oxygen (r = 0.71 p < 0.001; r = 0.74 p < 0.001; r = 0.70 p < 0.001; for left kidney, right kidney, and thigh NIRS respectively). There was a strong correlation between both kidney NIRS signals and thigh NIRS measurements (r = 0.85 p < 0.001; r = 0.88 p < 0.001;for left kidney vs thigh and right kidney vs thigh respectively). There was also a strong correlation between left and right kidney NIRS (r = 0.90 p < 0.001). These relationships were maintained regardless of the resuscitation protocol. These results suggest that kidney NIRS measurements were more closely related to thigh NIRS measurements than invasive kidney tissue oxygen concentration.

Combining Machine Learning and Urine Oximetry: Towards an Intraoperative AKI Risk Prediction Algorithm.

Acute kidney injury (AKI) affects up to 50% of cardiac surgery patients. The definition of AKI is based on changes in serum creatinine relative to a baseline measurement or a decrease in urine output. These monitoring methods lead to a delayed diagnosis. Monitoring the partial pressure of oxygen in urine (PuO2) may provide a method to assess the patient's AKI risk status dynamically. This study aimed to assess the predictive capability of two machine learning algorithms for AKI in cardiac surgery patients. One algorithm incorporated a feature derived from PuO2 monitoring, while the other algorithm solely relied on preoperative risk factors. The hypothesis was that the model incorporating PuO2 information would exhibit a higher area under the receiver operator characteristic curve (AUROC). An automated forward variable selection method was used to identify the best preoperative features. The AUROC for individual features derived from the PuO2 monitor was used to pick the single best PuO2-based feature. The AUROC for the preoperative plus PuO2 model vs. the preoperative-only model was 0.78 vs. 0.66 (p-value < 0.01). In summary, a model that includes an intraoperative PuO2 feature better predicts AKI than one that only includes preoperative patient data.

The impact of urine flow on urine oxygen partial pressure monitoring during cardiac surgery.

PURPOSE: Urine oxygen partial pressure (PuO2) may be useful for assessing acute kidney injury (AKI) risk. The primary purpose of this study was to quantify the ability of a novel urinary oxygen monitoring system to make real-time PuO2 measurements intraoperatively which depends on adequate urine flow. We hypothesized that PuO2 data could be acquired with enough temporal resolution to provide real-time information in both AKI and non-AKI patients. METHODS: PuO2 and urine flow were analyzed in 86 cardiac surgery patients. PuO2 data associated with low (< 0.5 ml/kg/hr) or retrograde urine flow were discarded. Patients were excluded if > 70% of their data were discarded during the respective periods, i.e., during cardiopulmonary bypass (CPB), before CPB (pre-CPB), and after CPB (post-CPB). The length of intervals of discarded data were recorded for each patient. The median length of intervals of discarded data were compared between AKI and non-AKI patients and between surgical periods. RESULTS: There were more valid PuO2 data in CPB and post-CPB periods compared to the pre-CPB period (81% and 90% vs. 31% of patients included, respectively; p < 0.001 and p < 0.001). Most intervals of discarded data were < 3 minutes during CPB (96%) and post-CPB (98%). The median length was < 25 s during all periods and there was no significant difference in the group median length of discarded data intervals for AKI and non-AKI patients. CONCLUSIONS: PuO2 measurements were acquired with enough temporal resolution to demonstrate real-time PuO2 monitoring during CPB and the post-CPB period. GOV IDENTIFIER: NCT03335865, First Posted Date: Nov. 8th, 2017.

Noninvasive and Invasive Renal Hypoxia Monitoring in a Porcine Model of Hemorrhagic Shock.

Up to 50% of patients with trauma develop acute kidney injury (AKI), in part due to poor renal perfusion after severe blood loss. AKI is currently diagnosed based on a change in serum creatinine concentration from baseline or prolonged periods of decreased urine output. Unfortunately, baseline serum creatinine concentration data is unavailable in most patients with trauma, and current estimation methods are inaccurate. In addition, serum creatinine concentration may not change until 24-48 h after the injury. Lastly, oliguria must persist for a minimum of 6 h to diagnose AKI, making it impractical for early diagnosis. AKI diagnostic approaches available today are not useful for predicting risk during the resuscitation of patients with trauma. Studies suggest that urinary partial pressure of oxygen (PuO2) may be useful for assessing renal hypoxia. A monitor that connects the urinary catheter and the urine collection bag was developed to measure PuO2 noninvasively. The device incorporates an optical oxygen sensor that estimates PuO2 based on luminescence quenching principles. In addition, the device measures urinary flow and temperature, the latter to adjust for confounding effects of temperature changes. Urinary flow is measured to compensate for the effects of oxygen ingress during periods of low urine flow. This article describes a porcine model of hemorrhagic shock to study the relationship between noninvasive PuO2, renal hypoxia, and AKI development. A key element of the model is the ultrasound-guided surgical placement in the renal medulla of an oxygen probe, which is based on an unsheathed optical microfiber. PuO2 will also be measured in the bladder and compared to the kidney and noninvasive PuO2 measurements. This model can be used to test PuO2 as an early marker of AKI and assess PuO2 as a resuscitative endpoint after hemorrhage that is indicative of end-organ rather than systemic oxygenation.

The Intraoperative Assessment of Right Ventricular Function During Cardiac Surgery.

The importance of right ventricular (RV) dysfunction in patients undergoing cardiac surgery is well recognized. There is extensive literature regarding the accurate assessment of RV dysfunction with both echocardiography and hemodynamic data, but the majority of these studies are with transthoracic echocardiography (TTE) and in awake patients. Many of the tools used to assess the RV with TTE are angle-dependent and, therefore, may be inaccurate with transesophageal echocardiography (TEE). Very few of these modalities have been validated either with TEE or in patients under general anesthesia. The purpose of this review is to discuss the intraoperative tools available to the cardiac anesthesiologist for the assessment of RV function. The authors review the available literature surrounding intraoperative RV assessment, from subjective assessment to traditional objective tools that were developed for TTE and newer technology that can be adapted to both TTE and TEE. Future work should focus on whether or not these intraoperative RV assessment tools predict outcome after cardiac surgery.

Noninvasive Urine Oxygen Monitoring and the Risk of Acute Kidney Injury in Cardiac Surgery.

BACKGROUND: Acute kidney injury (AKI) is a common complication of cardiac surgery. An intraoperative monitor of kidney perfusion is needed to identify patients at risk for AKI. The authors created a noninvasive urinary oximeter that provides continuous measurements of urinary oxygen partial pressure and instantaneous urine flow. They hypothesized that intraoperative urinary oxygen partial pressure measurements are feasible with this prototype device and that low urinary oxygen partial pressure during cardiac surgery is associated with the subsequent development of AKI. METHODS: This was a prospective observational pilot study. Continuous urinary oxygen partial pressure and instantaneous urine flow were measured in 91 patients undergoing cardiac surgery using a novel device placed between the urinary catheter and collecting bag. Data were collected throughout the surgery and for 24 h postoperatively. Clinicians were blinded to the intraoperative urinary oxygen partial pressure and instantaneous flow data. Patients were then followed postoperatively, and the incidence of AKI was compared to urinary oxygen partial pressure measurements. RESULTS: Intraoperative urinary oxygen partial pressure measurements were feasible in 86/91 (95%) of patients. When urinary oxygen partial pressure data were filtered for valid urine flows greater than 0.5 ml · kg-1 · h-1, then 70/86 (81%) and 77/86 (90%) of patients in the cardiopulmonary bypass (CPB) and post-CPB periods, respectively, were included in the analysis. Mean urinary oxygen partial pressure in the post-CPB period was significantly lower in patients who subsequently developed AKI than in those who did not (mean difference, 6 mmHg; 95% CI, 0 to 11; P = 0.038). In a multivariable analysis, mean urinary oxygen partial pressure during the post-CPB period remained an independent risk factor for AKI (relative risk, 0.82; 95% CI, 0.71 to 0.95; P = 0.009 for every 10-mmHg increase in mean urinary oxygen partial pressure). CONCLUSIONS: Low urinary oxygen partial pressures after CPB may be associated with the subsequent development of AKI after cardiac surgery.

Intraoperative Urinary Biomarkers and Acute Kidney Injury After Cardiac Surgery.

OBJECTIVES: To evaluate the association of intraoperative urinary biomarker excretion during cardiac surgery and the subsequent development of acute kidney injury (AKI). DESIGN: Prospective, nonrandomized, observational study. SETTING: Single tertiary-level, university-affiliated hospital. PARTICIPANTS: Ninety patients undergoing cardiac surgery with cardiopulmonary bypass (CPB). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Urinary samples were collected every 30 minutes intraoperatively and then at four, 12, and 24 hours after CPB. Samples were measured for interleukin 18 (IL-18), kidney injury molecule-1 (KIM1), and creatinine concentrations. Urinary biomarker excretion (raw and indexed to creatinine) for four intraoperative and three postoperative points were compared between patients with and those without subsequent AKI defined by increased serum creatinine concentration ≥0.3 mg/dL within the first 48 hours or ≥1.5 times baseline within seven days. Raw and indexed median IL-18 values were similar between AKI groups at all intraoperative points, but became significantly different at 12 hours after CPB. Raw and indexed median KIM1 values were significantly different between AKI groups at multiple intraoperative points and at four and 12 hours after CPB. During intraoperative and postoperative points, patients in the fourth quartile of KIM1 excretion had greater AKI incidence and longer intensive care and hospital lengths of stay than those in the first quartile. Only postoperatively did the differences in these outcomes between the fourth and first quartile of IL-18 excretion occur. CONCLUSIONS: Intraoperative KIM1 but not IL-18 excretion was associated with postoperative development of AKI.

Preventing or Minimizing Acute Kidney Injury in Patients Undergoing Transcatheter Aortic Valve Replacement.

BACKGROUND: Transcatheter aortic valve implantation (TAVI) is now routinely performed in patients with aortic stenosis with low mortality and complication rates. Although periprocedural risks have been substantially minimized, procedure- and contrast-induced acute kidney injury (AKI) remains a major concern. AKI remains a frequent complication of contrast-guided interventional procedures and is associated with a significantly adverse prognosis. We review the currently available clinical data related to AKI, with emphasis on contrast-induced nephropathy (CIN), and discuss a novel, integrated approach aiming to minimize AKI risk in high-risk patients. A stepwise algorithm is also proposed for the management of these complex patients.

The Year in Perioperative Echocardiography: Selected Highlights From 2019.

This article is the fourth of an annual series reviewing the research highlights of the year pertaining to the subspecialty of perioperative echocardiography for the Journal of Cardiothoracic and Vascular Anesthesia. The authors thank the editor-in-chief, Dr. Kaplan, and the editorial board, for the opportunity to continue this series. In most cases, these were research articles that were targeted at the perioperative echocardiography diagnosis and treatment of patients after cardiothoracic surgery; but in some cases, these articles targetted the use of perioperative echocardiography in general.

Strain Imaging: An Everyday Tool for the Perioperative Echocardiographer.

Strain analysis allows for global and regional analysis of myocardial function and has been shown to be an independent predictor of outcomes after cardiac surgery. Strain imaging offers advantages over traditional EF measurements in that it is relatively angle independent, it is less dependent upon loading conditions, it is reproducible, it does not rely on geometric assumptions, and it can detect subclinical systolic dysfunction. Limitations of strain analysis include high temporal resolution requirements, a strong dependence on image quality, and inter-vendor variability. In addition, there is a paucity of data on the intraoperative applications of strain. The ASE has defined a global longitudinal strain of -20% measured by transthoracic echocardiography to be considered normal, with less negative values considered abnormal. Presently, there are no published guidelines on the normal values of strain with transesophageal echocardiography (TEE). However, multiple studies have shown that a reduction in intraoperative strain assessed with TEE has been shown to be an independent predictor of complications during cardiac surgery. Accordingly, further incorporation of intraoperative strain analysis with TEE could aid in prognostication for patients undergoing cardiac surgery. As perioperative strain analysis continues to advance, an understanding of these concepts is imperative for perioperative echocardiographers. It is the authors' goal to show that strain imaging can provide a reliable and objective measure that can be performed in real time to aid in decision-making and perioperative risk stratification.

Regional Versus Global Measurements of Right Ventricular Strain Performed in the Operating Room With Transesophageal Echocardiography.

OBJECTIVE: To compare regional and global measures of right ventricular (RV) strain in patients undergoing intraoperative transesophageal echocardiography (TEE). DESIGN: Prospective, nonrandomized, observational study. SETTING: Single tertiary-level, university hospital. PARTICIPANTS: The study comprised 48 patients undergoing intraoperative TEE. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A global RV strain measurement (termed RV 5-wall strain [RV 5WS]) was calculated by averaging the longitudinal strain calculated from multiple TEE views. This global strain measurement was compared with the more standard regional strain measurements obtained in a single 4-chamber view (RV free-wall strain [RV FWS] and RV global longitudinal strain [RV GLS]) and with traditional measures of RV function. Regional and global strain measurements were feasible in the operating room. RV FWS and RV GLS strongly correlated with RV 5WS (r = 0.86 and 0.87, respectively) with no significant bias and limits of agreement of approximately -5% to 5%. RV FWS and RV GLS were even more closely correlated with each other (r = 0.99) with no significant bias and limits of agreement less than -2% to 2%. Both regional and global RV strain measurements showed a high sensitivity (RV FWS 94%; RV GLS 94%; RV 5WS 89%) and moderate specificity (RV FWS 70%; RV GLS 67%; RV 5WS 63%) for RV dysfunction based on a reference standard of 3-dimensional RV ejection fraction. CONCLUSIONS: Both regional and global RV strain measurements are feasible in the operating room with TEE. Regional and global measures of RV function correlate well and are sensitive indicators of RV dysfunction.

A Comparison of Left- and Right-Sided Strain Software for the Assessment of Intraoperative Right Ventricular Function.

OBJECTIVE: To compare intraoperative right ventricular (RV) strain measurements made with left ventricular (LV) strain software commonly found on the echocardiography machine (Philips QLAB chamber motion quantification, version 10.7, Philips, Amsterdam, The Netherlands), with offline analysis using the dedicated RV strain software (EchoInsight, version 2.2.6.2230, Epsilon Imaging, Ann Arbor, MI). DESIGN: Prospective, nonrandomized, observational study. SETTING: Single tertiary level, university-affiliated hospital. PARTICIPANTS: The study comprised 48 patients undergoing transesophageal echocardiography for cardiac or noncardiac surgery. INTERVENTIONS: Two-dimensional (2D) and 3-dimensional (3D) images of the right ventricle were obtained. Intraoperative 2D images were analyzed in real time for RV free wall strain (FWS) and global longitudinal strain (GLS) using QLAB chamber motion quantification (CMQ) LV strain software on the echocardiography machine. Two dimensional images were then analyzed offline to determine the RV FWS and GLS using EchoInsight RV-specific strain software. Three-dimensional images were then analyzed offline to detemine the 3D RV ejection fraction (3D RV EF) using TomTec 4D RV function (Unterschleissheim, Germany). Spearman's correlation and Bland-Altman analyses were used to characterize the relationship between RV strain measurements. Both types of strain measurements were compared to a reference standard of 3D RV EF. MEASUREMENTS AND MAIN RESULTS: Intraoperative RV strain measurements using LV-specific strain software correlated with offline RV strain measurements using the RV-specific strain software (FWS rho = 0.85; GLS rho = 0.81). The bias and limits of agreement were 0.75% (- 6.66 to 8.17) for FWS and -4.53% (-11.55 to 2.50) for GLS. The sensitivity and specificity for RV dysfunction for the intraoperative LV-specific software were 94% (95% confidence interval [CI] 73-100) and 70% (95% CI 51-85), respectively, for RV FWS and 94% (95% CI 73-100) and 67% (95% CI 47-83), respectively, for RV GLS. The sensitivity and specificity for RV dysfunction for the offline RV-specific software were 89% (95% CI 65-99) and 73% (95% CI 54-88), respectively, for RV FWS and 94% (95% CI 73-100) and 30% (95% CI 15-49), respectively, for RV GLS. CONCLUSION: Intraoperative RV strain measurements using LV-specific strain software commonly available on the echocardiography machine (QLAB CMQ) correlate with offline RV strain measurements using RV-specific strain software (EchoInsight). The bias and limits of agreement for these left- and right-sided strain software suggest that these 2 measures of RV function cannot be used interchangeably. Both, however, were sensitive measures of RV dysfunction and therefore are likely clinically relevant.

Intraoperative Transesophageal Echocardiography and Right Ventricular Failure After Left Ventricular Assist Device Implantation.

OBJECTIVE: To determine whether intraoperative measures of right ventricular (RV) function using transesophageal echocardiography are associated with subsequent RV failure after left ventricular assist device (LVAD) implantation. DESIGN: Retrospective, nonrandomized, observational study. SETTING: Single tertiary-level, university-affiliated hospital. PARTICIPANTS: The study comprised 100 patients with systolic heart failure undergoing elective LVAD implantation. INTERVENTIONS: Transesophageal echocardiographic images before and after cardiopulmonary bypass were analyzed to quantify RV function using tricuspid annular plane systolic excursion (TAPSE), tricuspid annular systolic velocity (S'), fractional area change (FAC), RV global longitudinal strain, and RV free wall strain. A chart review was performed to determine which patients subsequently developed RV failure (right ventricular assist device placement or prolonged inotrope requirement ≥14 days). MEASUREMENTS AND MAIN RESULTS: Nineteen patients (19%) subsequently developed RV failure. Postbypass FAC was the only measure of RV function that distinguished between the RV failure and non-RV failure groups (21.2% v 26.5%; p = 0.04). The sensitivity, specificity, and area under the curve of an abnormal RV FAC (<35%) for RV failure after LVAD implantation were 84%, 20%, and 0.52, respectively. No other intraoperative measure of RV function was associated with subsequent RV failure. RV failure increased ventilator time, intensive care unit and hospital length of stay, and mortality. CONCLUSION: Intraoperative measures of RV function such as tricuspid annular plane systolic excursion, tricuspid annular systolic velocity, and RV strain were not associated with RV failure after LVAD implantation. Decreased postbypass FAC was significantly associated with RV failure but showed poor discrimination.