Intraoperative monitor of aggregate organ injury time

Abstract

A computer system and a computer readable medium are disclosed that are configured to perform and store a computer-implemented method for reducing acute kidney injury during cardiac surgery. The method includes measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP); storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury.

Description

INTRAOPERATIVE MONITOR OF AGGREGATE ORGAN INJURY TIMECross Reference to Related Applications

[0001] This application claims priority to U.S. Provisional Application No. 63 / 622,315 filed on January 18, 2024, the contents of which are hereby incorporated by reference in its entirety.Background

[0002] Cardiac surgery associated acute kidney injury (CSA-AKI) is common and attributed, in part, to hemodynamic parameters related to kidney perfusion. Clinical guidelines and standards of care recommend clinicians target mean arterial pressure (MAP) and central venous pressure (CVP) to specific ranges to reduce risk of CSA-AKI. Reviewed in this application is the relation of CSA-AKI with time spent in a full set of clinically relevant ranges for MAP and CVP, as they occurred separately and concurrently, during cardiac surgery.

[0003] Acute kidney injury (AKI) is a common, costly, deadly problem in cardiac surgery even for people that present to surgery with normal kidneys. Low blood flow to the kidneys is called hypoperfusion and is a common cause of AKI. Hypoperfusion is determined by 3 monitored variables: mean arterial pressure, central venous pressure, and cardiac output. Cardiac output may be calculated from analysis of the patient’s arterial waveform, e.g., using the FDA approved Retia Monitor.Summary

[0004] According to examples of the present disclosure, a computer-implemented method for reducing acute kidney injury during cardiac surgery is disclosed. The method comprises measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP); storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury. In some examples, the method can include one or more of the following features. The MAP data range is 45-115 mmHg and the CVP data range is 0-20 mmHg. The MAP and the CVP are measured andrecorded at a resolution of no greater than 1 minute resolution. The MAP and the CVP are measured and recorded from an intraoperative electronic health record. The first pressure window is 5 mmHg and the second pressure window is 2 mmHg.

[0005] According to examples of the present disclosure, a computer system is disclosed that comprises a hardware processor; a non-transitory computer-readable medium that stores instruction, that when executed by the hardware processor, perform a method for reducing acute kidney injury during cardiac surgery, comprising: measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP); storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury. In some examples, the method can include one or more of the following features. The MAP data range is 45-115 mmHg and the CVP data range is 0-20 mmHg. The MAP and the CVP are measured and recorded at a resolution of no greater than 1 minute resolution. The MAP and the CVP are measured and recorded from an intraoperative electronic health record. The first pressure window is 5 mmHg and the second pressure window is 2 mmHg.

[0006] According to examples of the present disclosure, a non-transitory computer-readable medium that stores instruction, that when executed by a hardware processor, perform a method for reducing acute kidney injury during cardiac surgery, comprising: measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP); storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury. In some examples, the method can include one or more of the following features. The MAP data range is 45-115 mmHg and the CVP data range is 0-20 mmHg. The MAP and the CVP are measured and recorded at a resolution of nogreater than 1 minute resolution. The MAP and the CVP are measured and recorded from an intraoperative electronic health record. The first pressure window is 5 mmHg and the second pressure window is 2 mmHg.Brief Description of the Drawings

[0007] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present teachings and together with the description, serve to explain the principles of the present teachings. In the figures:

[0008] FIG. 1 illustrates a patient exclusion and inclusion flow diagram according to examples of the present disclosure.

[0009] FIG. 2A illustrates patient exposure to mean arterial pressure ranges; the mean total time spent in each MAP range (minutes) is displayed with one standard deviation for the whole population alongside the percent of individuals, in red, that spent at least 5 minutes within each range according to examples of the present disclosure.

[0010] FIG. 2B illustrates the association between each 5 minutes spent in range with acute kidney injury; the association between MAP and AKI across all MAP ranges in three models, i.e., the unadjusted, adjusted for all table 1 covariates, and the adjusted with 95% CI adjusted for multiplicity of comparisons; brackets denote numbers included in range while parentheses denote numbers up to the integer according to examples of the present disclosure.

[0011] FIG. 3A illustrates the summary of exposure to central venous pressure ranges; mean total time spent in each CVP range (minutes) is displayed with one standard deviation (error bar) for the whole population alongside the percent of individuals, in red, that spent at least 5 minutes within each range according to examples of the present disclosure.

[0012] FIG. 3B illustrates the association between each 5 minutes spent in range with acute kidney injury; three models are shown, the unadjusted, adjusted for all table 1 covariates, and the adjusted with 95% CI adjusted for multiplicity of comparisons; a J shaped curve is visualized with lowest odds of AKI within a CVP range of 4-6 mmHg; brackets denote numbers included in range while parentheses denote numbers up to the integer according to examples of the present disclosure.

[0013] FIG. 4A and FIG. 4B show the odds ratio point estimates are displayed for each range and are color coded by the following scale (darker shade: above OR=1 indicating higher odds of AKI vs lighter shade: less than OR=1 indicating lower odds of AKI) according to examples of the present disclosure. Darker scales are used to demonstrate the strength of OR. An asterisk denotes p value below 0.05. A nonlinear interaction is visualized between MAP, CVP, andAKI. Odds of AKI appear lowest in the right lower corner (darker shade) with odds of AKI increasing towards the upper left comer. Brackets denote numbers included in range while parentheses denote numbers up to the integer.

[0014] FIG. 5 is an illustration of a Bowman’s capsule of a human kidney.

[0015] FIG. 6A and FIG. 6B illustrate odds ratios for results of cardiac surgery study according to examples of the present disclosure.

[0016] FIG. 7A and FIG. 7B are more generalized versions of FIG. 6A and FIG. 6B with various odds ratios grouped by outcome according to examples of the present disclosure.

[0017] FIG. 8A and FIG. 8B illustrates results of study of low CO across different MAP thresholds and associated renal injury according to examples of the present disclosure.

[0018] FIG. 9 shows various data collected during studies, and in particular, a summary of exposure within concurrent MAP / CVP ranges for number of patients are the population with greater than 5 minutes within range according to examples of the present disclosure.

[0019] FIG. 10 shows various data collected during studies, and in particular, a summary of exposure within concurrent MAP / CVP ranges with the mean (standard deviation) minutes for each range according to examples of the present disclosure.

[0020] FIG. 11 shows various data collected during studies, and in particular, a summary of association data between 5 minutes spent in concurrent MAP / CVP ranges and secondary outcome acute kidney injury 48 hours adjusted for all covariates according to examples of the present disclosure. Time spent within the concurrent MAP / CVP ranges were individually regressed on the outcome AKI. The Odds ratio point estimates are displayed for each range and are color coded by the following scale (darker shade: above OR=1 indicating higher odds of AKI v ligher shade: less than OR=1 indicating lower odds of AKI). Darker scales are used to demonstrate the strength of OR. Ranges in gray (b) had insufficient observations for full adjustment. An asterix denotes p value< 05. The 95% CI are also displayed under the OR point estimate. A nonlinear interaction is visualized between MAP, CVP, and AKI. Odds of AKI appear lowest in the right lower corner (scale 1102, 1104) with odds of AKI increasing towards the upper left comer.

[0021] FIG. 12 shows various data collected during studies, and in particular, a summary of association data between 5 minutes spent in concurrent MAP / CVP ranges and secondary outcome acute kidney injury 48 hours adjusted for all covariates and for multiple comparisons according to examples of the present disclosure. From FIG. 10, the 95%CI of each MAP / CVP are adjusted for multiple comparisons using the CMA method.

[0022] FIG. 13 A and FIG. 13B show various data collected during studies, and in particular, a summary of association between 5 minutes spent in concurrent MAP / CVP ranges and secondary acute kidney injury outcome (FIG. 13 A) unadjusted and (FIG. 13 A) adjusted for all covariates according to examples of the present disclosure. Time spent within the concurrent MAP / CVP ranges were individually regressed on the outcome AKI. The Odds ratio point estimates are displayed for each range and are coded by the following scale (red: above OR=1 indicating higher odds of AKI vs scale 1304, 1306, 1308, 1310: less than OR=1 indicating lower odds of AKI). Darker scales are used to demonstrate the strength of OR. Ranges in gray (FIG. 13B) had insufficient observations for full adjustment. An asterix denotes p value< 05. The 95% CI are also displayed under the OR point estimate. A nonlinear interaction is visualized between MAP, CVP, and AKI. Odds of AKI appear lowest in the right lower corner (scale 1322, 1326) with odds of AKI increasing towards the upper left comer.

[0023] FIG. 14 shows various data collected during studies, and in particular, a summary of association data between 5 minutes spent in concurrent MAP / CVP ranges and secondary acute kidney injury outcome adjusted for all covariates and for multiple comparisons according to examples of the present disclosure. From FIG. 13B, the 95%CI of each MAP / CVP are adjusted for multiple comparisons using the CMA method.

[0024] FIG. 15A and FIG. 15B show various data collected during studies, and in particular, a summary of identified 5 zones of Mean Arterial Pressure and Central Venous Pressure evaluated together with either increased risk (Zones 4 and 5) or protection (zones 1 and 2) from AKI from the data in FIG. 14 according to examples of the present disclosure.

[0025] FIG. 16 shows plots of the relationship between Zones 4 / 5 on AKI probability from the data in FIG. 14 according to examples of the present disclosure.

[0026] FIG. 17 shows plots of the relationship of Zone 3 on AKI probability from the data in FIG. 14 according to examples of the present disclosure.

[0027] FIG. 18 shows plots of the relationship of Zone 1 and Zone 2 exposure vs Zone 1 and Zone 2 on AKI probability from the data in FIG. 14 according to examples of the present disclosure.

[0028] FIG. 19 shows plots of the exposure during the different time phases of surgery that contribute to AKI probability from the data in FIG. 14 according to examples of the present disclosure.

[0029] FIG. 20 show plots of trajectory of probability (AKI) from the data in FIG. 14 according to examples of the present disclosure.

[0030] FIG. 21 show plots of Harm Accumulation Function versus duration in minutes for different adjected K values according to examples of the present disclosure.

[0031] FIG. 22 shows a plot of the summary of exposure to narrow hemodynamic ranges for mean arterial pressure (MAP) according to examples of the present disclosure. Mean cumulative time (SD) spent in each MAP range in minutes (lower bars); percent of individuals with at least 5 minutes in each range (higher bars).

[0032] FIG. 23 and FIG. 24 show plots of the sensitivity analyses for associations of CSA-AKI with a series of MAP ranges in increments of 10 and 15 mm Hg, respectively, according to examples of the present disclosure. Odds ratios and 95% Cis shown from each regression model.

[0033] FIG. 25 shows a plot of the sensitivity analyses for associations of AKI with MAP ranges stratified by phase of surgery according to examples of the present disclosure. Odds ratios and 95% Cis shown for each regression model adjusted for all covariates from the primary analysis. Odds ratios not shown when data was too sparse to generate point estimates for risk.

[0034] FIG. 26 shows a plot of the summary of exposure to narrow hemodynamic ranges for central venous pressure (CVP) according to examples of the present disclosure. Mean cumulative time (SD) spent in each CVP range in minutes (lower bars); percent of individuals with at least 5 minutes in each range (higher bars).

[0035] FIG. 27 shows a plot of the sensitivity analyses for association of AKI with CVP ranges stratified by phase of surgery according to examples of the present disclosure. Odds ratios and 95% Cis shown for each regression model adjusted for all covariates from the primary analysis.

[0036] FIG. 28 and FIG. 29 show the summary of Exposure within Joint Concurrent MAP / CVP ranges, according to examples of the present disclosure, where FIG. 28 shows Mean (SD) cumulative minutes for each range and FIG. 29 shows number of subjects (%) with at least 5 minutes within range.

[0037] FIG. 30 and FIG. 31 show data of the association of acute kidney injury (AKI) with cumulative time spent in narrow ranges of joint concurrent MAP / CVP according to examples of the present disclosure. MAP and CVP were examined concurrently at 1 -minute intervals. We determined the cumulative time when both MAP and CVP were within each of 70 joint MAP / CVP ranges. Separate logistic regression models estimated odds ratios (OR) and 95% confidence intervals for AKI within each joint MAP / CVP range. ORs are expressed per five minutes in range. ORs >=1.03 are shaded 3108, 3112, ORs<=0.97 are shaded 3102, 3106, withintensity of color denoting increasing effect size. Ranges shaded in gray indicate that data were too sparse to have confidence in point estimates after adjustment for covariates. An asterisk denotes p <.05. (a) ORs / 95% Cis in each of the 70 joint MAP / CVP ranges adjusted for all covariates, (b) ORs / 95% Cis in each of the 70 joint MAP / CVP ranges adjusted for all covariates plus adjustments for correlation of data and multiple comparisons.

[0038] FIG. 32 shows computer-generated displays for a patient that were generated using the methods described herein according to examples of the present disclosure.

[0039] FIG. 33 shows a flow chart for a computer-implemented method for reducing acute kidney injury during cardiac surgery.

[0040] FIG. 34 shows a computer system according to examples of the present disclosure.Detailed Description

[0041] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings and figures. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to one of ordinary skill in the art that the invention may be practiced without these specific details. In other instances, well-known methods, procedures, components, circuits, and networks have not been described in detail so as not to unnecessarily obscure aspects of the embodiments.

[0042] It will also be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first object or step could be termed a second object or step, and, similarly, a second object or step could be termed a first object or step, without departing from the scope of the present disclosure. The first object or step, and the second object or step, are both, objects or steps, respectively, but they are not to be considered the same object or step.

[0043] The terminology used in the description herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used in this description and the appended claims, the singular forms “a,” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term “and / or” as used herein refers to and encompasses any possible combinations of one or more of the associated listed items. It will be further understood that the terms “includes,” “including,” “comprises” and / or “comprising,” when used in this specification, specify thepresence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Further, as used herein, the term “if’ may be construed to mean “when” or “upon” or “in response to determining” or “in response to detecting,” depending on the context.

[0044] Cardiac Surgery Associated Acute Kidney Injury (CSA-AKI) is common and related to both short- and long-term chronic kidney disease and mortality. Intraoperative hemodynamic perturbations, most notably those related to arterial hypotension and use of cardiopulmonary bypass (CPB), are thought to be important causes of CSA-AKI. The number of minutes exposed to MAP below 65 mmHg is reported to be an important risk factor for development of CSA-AKI. A MAP below 65 mmHg is often used as the threshold to define hypotension. Venous congestion is also a potential contributor to CSA-AKI but is less well studied in cardiac surgery. Recent reports suggest that elevated central venous pressure (CVP) is associated with CSA-AKI, and that venous congestion may be a more important contributor to CSA-AKI than arterial hypotension. Based on low to moderate evidence, current clinical guidelines in cardiac surgery recommend maintaining MAP greater than 65 mmHg and CVP of 8-10 mmHg.

[0045] Hemodynamic management strategies have garnered substantial interest for prevention of CSA-AKI because unlike many AKI risk factors such as age, chronic kidney disease, ejection fraction, duration of CPB, etc., hemodynamic parameters can be modified. Although several studies have converged on a lower MAP threshold of 65 mmHg in adults, it is unclear to what degree venous congestion contributes to CSA-AKI, and what threshold for CVP, if any, might exist. Importantly, blood pressure and venous pressure are concurrent and interacting physiologic parameters, and no study has investigated the association of AKI with these hemodynamic parameters as they occur concurrently in vivo.

[0046] A retrospective cohort analysis of patients undergoing coronary artery bypass (CAB) surgery was conducted to evaluate the association of CSA-AKI with MAP and CVP, separately and as they occur concurrently, in vivo. It was hypothesized that arterial hypotension (defined by MAP) and venous congestion (defined by CVP), separately and concurrently, would be related to CSA-AKI, and the nature of the relation between these hemodynamic parameters and CSA-AKI through the range of values that occurs during cardiac surgery was sought. Because of the inter-relatedness of MAP and CVP and correlation of these time-series hemodynamic measurements, a novel time-in-range regression approach to examine the associations of hemodynamic parameters with CSA-AKI was used. This novel methodologyallowed nonlinear nonparametric associations to be addressed and adjustment for multiple comparisons, providing clearer insight into the relation of hemodynamic parameters with CSA- AKI and potential threshold effects.MethodsData Sources

[0047] Subjects were eligible for inclusion if they underwent isolated CAB requiring CPB at the Johns Hopkins Hospital between July 1, 2016 to October 31, 2019. Subjects were excluded sequentially for the following: 1) end stage renal disease or baseline serum creatinine above 4 mg / dL; 2) surgical procedure required more than 1 aortic cross clamp; 3) intraoperative use of extracorporeal membrane oxygenation; 4) multiple surgical procedures during the same hospitalization; and, 5) hemodynamic (MAP, CVP) data in the electronic health record (EHR) were insufficient for analysis. Eligible subjects were identified from the Johns Hopkins Society of Thoracic Surgeons (STS) registry. Data elements from the STS were merged with data extracted from the EHR into a single dataset.Primary Exposures: Pressure ranges for MAP, CVP, and concurrent MAP and CVP

[0048] Data for MAP and CVP were extracted at 1 minute resolution from the intraoperative electronic health record. Preprocessing of raw time series data, including strategies to manage artifact, outliers, and missing data, was conducted. MAP values were divided into increments of 5 mmHg through the range of values from 45-115 mmHg, and CVP ranges were divided in increments of 2 mmHg through the range of values from 0-20 mmHg. Note that the term “MAP G [45,55]” means the set of mean arterial pressure results closest to 45 mmHg and 55 mmHg and “CVP E [0,2]” means the set of central venous pressure closes to 0 mmHg and 2 mmHg. To model concurrent MAP and CVP, the ranges depend on the number of minutes in range for both MAP and CVP simultaneously. More precisely, 70 bivariate ranges were created from the 7 univariate MAP and 10 CVP increments. For example, the first bivariate range was MAP E [45,55] and CVP G [0,2], the second MAP 6 [45,55] and CVP G (2,4], and the last was MAP G (105, 115] and CVP G (18,20], For MAP, there were only 7 intervals instead of 14 five-unit intervals because 10-unit increments in MAP were used to ensure that enough data were available within the bivariate ranges.Acute Kidney Injury Outcome (AKI) and Covariates

[0049] The primary outcome was AKI 48 hours after the end of surgery as defined by Kidney Disease Improving Global Practice (KDIGO) criteria using change in serum creatinine, i.e., an increase in serum creatinine greater than or equal to 0.3 mg / dl from baseline. As a secondaryoutcome, AKI by KDIGO criteria at 48 hours or within 7 days of surgery using change in serum creatinine was used, i.e., using the 48-hour criteria or increase in serum creatinine greater than or equal to 1.5 times the baseline within 7 days after surgery. As in recent studies, KDIGO urine output criteria was not included due to potential confounding by diuretic administration. The following baseline covariates were examined and included in adjusted analyses: age, male / female, BMI, hypertension, diabetes, prior myocardial infarction, congestive heart failure, left ventricular ejection fraction, intra-aortic balloon pump, stroke, peripheral vascular disease, chronic lung disease, use of beta blockers, use of ACE inhibitors or ARBs, use of statins, preoperative hematocrit, baseline creatinine, redo sternotomy, emergency case, and STS risk of mortality.

[0050] The following intra-operative covariates were also examined and included in adjusted models: length of procedure, length of CPB, cumulative vasopressor inotrope dose weighted across medications, total crystalloid administered, and total packed red blood cell transfusions. Statistical Analysis

[0051] To model the association between MAP (CVP) and AKI, a time-in-range regression approach was implemented. Logistic regression was used to quantify the association between AKI and each individual time in range for MAP and CVP. The models for MAP had the following structure:AKI|MAP range, ; (1)AKI|MAP range, +Covariates . (2)

[0052] Model (1) was unadjusted for confounders while model (2) is adjusted for all covariates. A similar set of models was considered for CVP, though the number of ranges (and models) was smaller. Correlation and multiplicity adjusted (CMA) confidence intervals were obtained to account for the correlation between the tests and for the number of tests. The Bonferroni correction was also used to provide conservative multiplicity adjusted confidence intervals that do not account for correlation. CMA was developed to assess the possibility of Type I error in the setting of many comparisons that have a dependence structure.

[0053] To model the joint association between MAP, CVP, and AKI, similar time-in-range logistic regressions were conducted. We report univariate analyses for each concurrent MAP / CVP range based on the following model:AKI|MAP,CVP range? (3)Report is made of covariate adjusted estimates for MAP / CVP only within ranges with sufficient cases and controls data.

[0001] Sample size was determined by including every available patient within the study time period. Significance level p-value of less than 0.05 was designated for statistical significance, and CMA, described above, was used to address the possibility of Type I error in multiple comparisons. All analyses were performed using R studio.Results

[0002] FIG. 1 illustrates a patient exclusion and inclusion flow diagram according to examples of the present disclosure. A total of 1372 individuals underwent isolated CAB and 1199 met the inclusion criteria (FIG. 1). Characteristics of the study sample with and without AKI are shown in Table 1.

[0003] Table 1Of the 1199 subjects, 338 (28%) developed AKI within 48 hours after surgery. As expected, subjects who developed AKI differed from those who did not. Of particular note, subjects who developed AKI at 48 hours were slightly older, were more likely to have several comorbidities, including hypertension, congestive heart failure, and chronic lung disease, and had lower hematocrit and higher creatinine at baseline. Those who developed AKI also had higher predicted mortality by STS risk score, longer surgical and CPB times, and were more likely to receive packed red cell transfusion. Also examined were baseline characteristics by tertiles of exposure to total minutes of MAP below 65 mmHg and, separately, total minutes of CVP>12mmHg. Time below MAP of 65 mmHg was associated with higher vasoactive inotrope dosing, longer durations of CPB and surgery, and more blood transfusions. Time above CVP of 12 mmHg was associated with higher BMI, more congestive heart failure and lung disease, and longer durations of CPB and surgery.Mean Arterial Pressure and CSA-AKI

[0004] FIG. 2A illustrates patient exposure to mean arterial pressure ranges; the mean total time spent in each MAP range (minutes) is displayed in lower bars with one standard deviation for the whole population alongside the percent of individuals, in red, that spent at least 5 minutes within each range according to examples of the present disclosure. FIG. 2B illustratesthe association between each 5 minutes spent in range with acute kidney injury; the association between MAP and AKI across all MAP ranges in three models, i.e., the unadjusted, adjusted for all table 1 covariates, and the adjusted with 95% CI adjusted for multiplicity of comparisons; brackets denote numbers included in range while parentheses denote numbers up to the integer according to examples of the present disclosure.

[0005] The proportion of subjects who spent at least 5 minutes and the total time spent in each MAP range are shown on FIG. 2 A. Nearly half of all subjects (46%) were exposed to 5 or more minutes in the lowest MAP range (45-50 mmHg), and nearly all subjects were exposed to 5 or more minutes in the 55-60 mmHg range. The average time spent in MAP range of 45-50 was 6.2 ± 7.0 min; and the average time spent in 55-60 range was 34 ± 23 minutes. The most common MAP range was 65- 70 mmHg with mean (SD) of 65 (28) minutes. The least common MAP range was 110-115 mmHg with mean (SD) time in range of 3.1 (3.8) minutes.

[0006] The association between time spent in each MAP range and AKI outcome is shown on FIG. 2B. MAP ranges <65 mmHg were associated with increased odds of AKI, with asignificant increase in odds observed for every 5 minutes spent in ranges 45-50 (OR 1.176 / 95% CI 1.025-1.349), 50-55 (OR 1.129 / 1.025-1.243), and 55-60 (OR 1.059 / 1.014-1.107) mmHg. Each 5 minutes spent in the MAP range of 90-95 was associated with a significant reduction in odds of AKI (OR 0.848 / 95% CI 0. 0.762-0.943). Visual inspection of point estimates for the association between AKI and blood pressure through the full set of MAP ranges suggests a nonlinear relation between time spent in each MAP range and AKI outcome, with optimal AKI outcome associated with MAP of 90-95 mmHg (FIG. 2B).Central Venous Pressure and CSA-AKI

[0060] FIG. 3A illustrates the summary of exposure to central venous pressure ranges; mean total time spent in each CVP range (minutes) is displayed in lower bars with one standard deviation (error bar) for the whole population alongside the percent of individuals, in red, that spent at least 5 minutes within each range according to examples of the present disclosure. FIG. 3B illustrates the association between each 5 minutes spent in range with acute kidney injury; three models are shown, the unadjusted, adjusted for all Table 1 covariates, and the adjusted with 95% CI adjusted for multiplicity of comparisons; a J shaped curve is visualized with lowest odds of AKI within a CVP range of 4-6 mmHg; brackets denote numbers included in range while parentheses denote numbers up to the integer according to examples of the present disclosure.

[0061] The proportion of subjects who spent at least 5 minutes and the total time spent in each CVP range are shown on FIG. 3 A. A large majority of subjects spent at least 5 minutes in CVP ranges below 6 mmHg and the majority spent at least 5 minutes in CVP ranges above 14 mmHg. The average time spent in the 4-6 range was 43 ± 32 minutes and the average time spent in the 16-18 range was 10 ± 16 minutes. The most common CVP range was 6- 8 mmHg with mean (SD) time in range of 48 (33) minutes.

[0062] The association between time spent in each CVP range and AKI outcome is shown on FIG. 3B. CVP ranges >8 mmHg were associated with higher odds of AKI, with a significantly increased risk of AKI observed for every 5 minutes spent in ranges of 12-14 (OR 1.038 / 95% CI 1.002-1.076), 14-16 (OR 1.046 / 1.001-1.093), and 16-18 (OR 1.071 / 1.011-1.135) mmHg in univariate and adjusted models. CVP range of 4-6 mmHg was associated with significantly lower odds of AKI in univariate (OR 0.979 / 0.960-0.999) and covariate adjusted models (OR 0.974 / 0.951-0.997); however, confidence intervals surpassed the 95% significance threshold (OR 0.974 / 0.944-1.004) after multiplicity correction. Visual inspection of point estimates for the association between AKI and CVP through the full set of ranges suggests a nonlinearrelation between time spent in each CVP range and AKI outcome, with optimal AKI outcome at a CVP of 4-6 (FIG. 3B).Concurrent Mean Arterial Pressure / Central Venous Pressure and CSA-AKI

[0063] FIG. 10 shows various data collected during studies, and in particular, a summary of exposure within concurrent MAP / CVP ranges for number of patients are the population with greater than 5 minutes within range according to examples of the present disclosure. FIG. 11 shows various data collected during studies, and in particular, a summary of exposure within concurrent MAP / CVP ranges with the mean (standard deviation) minutes for each range according to examples of the present disclosure.

[0064] Because blood flow to any tissue is proportional to the difference between MAP and the downstream venous pressure and because this pressure differential is physiologically relevant at each instantaneous moment in time, we generated ranges for MAP and CVP as they occurred simultaneously in vivo. For these analyses, we divided MAP by 10 mmHg increments and CVP by 2 mmHg increments and determined the number of minutes in which each MAP and CVP value were present concurrently for each subject for every minute of the intraoperative period. The number (%) of subjects exposed to at least 5 minutes and the average (SD) number of minutes in each concurrent MAP / CVP range are shown in FIG. 10 and FIG. 11. As expected, the proportion of subjects and average time spent was at a minimum at extremes of concurrent MAP / CVP ranges. Least common was MAP 105-115 across all CVP ranges and CVP 18-20 across all MAP ranges. MAP 45-55 mmHg / CVP 18-20 mmHg was the least common concurrent range with only 8 individuals registering 5 minutes or more. The most common range was MAP 65-75 mmHg / CVP 6-8 mmHg- 80% of subjects spent at least 5 minutes in that range with mean (SD) of 17 (14) minutes.

[0065] FIG. 4A and FIG. 4B show the odds ratio point estimates are displayed for each range and are color coded by the following scale (above OR=1 indicating higher odds of AKI vs less than OR=1 indicating lower odds of AKI) according to examples of the present disclosure. Scales are used to demonstrate the strength of OR, such as 402, 404, 406, 408, 410, 412, 414, 416, 418, and 420 and scales 422, 424, 426, 428, 430, 432, 434, 436, 438, and 440. An asterisk denotes p value below 0.05. A nonlinear interaction is visualized between MAP, CVP, and AKI. Odds of AKI appear lowest in the right lower comer with odds of AKI increasing towards the upper left comer. Brackets denote numbers included in range while parentheses denote numbers up to the integer.

[0066] The association between AKI and each concurrent MAP / CVP range is shown on FIG. 4A and FIG. 4B. Odds of AKI were significantly increased for concurrent ranges of MAP 55- 85 / CVP 12-18 in both univariate and multivariable adjusted models. Odds of AKI were significantly reduced for concurrent ranges of MAP 65-105 / CVP 4-8 in both univariate and multivariable adjusted models. Odds of AKI was even higher at low extremes of MAP / high extremes of CVP and even lower at high extremes of MAP / low extremes of CVP; however, data were too sparse to have confidence in these point estimates. After multiplicity correction for all 70 bivariate MAP / CVP ranges, point estimates for risk did not change; however, confidence intervals exceeded the 95% significance threshold for each of the individual MAP / CVP ranges. Visual inspection of data showing associations of AKI through the full range of concurrent MAP / CVP ranges suggests a nonlinear relation between time spent in each concurrent MAP / CVP range and AKI outcome. Overall, there appeared to be a gradient of risk that was lowest at the highest MAPs / lowest CVPs (lower right comer of FIG. 4 A) and highest at lowest MAPs / highest CVPs (upper left corner of FIG. 4 A).

[0067] FIG. 5 is an illustration of a Bowman’s capsule of a human kidney.

[0068] FIG. 6A and FIG. 6B illustrate odds ratios for results of cardiac surgery study according to examples of the present disclosure, with scale 602, 604, 606, 608, and 610 shown.

[0069] FIG. 7A and FIG. 7B are more generalized versions of FIG. 6A and FIG. 6B with various odds ratios grouped by outcome according to examples of the present disclosure.

[0070] FIG. 8A and FIG. 8B illustrates results of study of low CO across different MAP thresholds and associated renal injury according to examples of the present disclosure, with scales 802, 804, 806, 808, and 810 and 812, 814, 816, 818, and 820, as shown.

[0071] FIG. 9 shows various data collected during studies, and in particular, a summary of exposure within concurrent MAP / CVP ranges for number of patients are the population with greater than 5 minutes within range according to examples of the present disclosure.

[0072] FIG. 10 shows various data collected during studies, and in particular, a summary of exposure within concurrent MAP / CVP ranges with the mean (standard deviation) minutes for each range according to examples of the present disclosure.

[0073] Secondary Outcome: AKI within 7 days after surgery

[0074] FIG. 11 shows various data collected during studies, and in particular, a summary of association data between 5 minutes spent in concurrent MAP / CVP ranges and secondary outcome acute kidney injury 48 hours adjusted for all covariates according to examples of the present disclosure, with scales 1102, 1104, 1106, 1108, 1110, 1112, 1114, 1116, and 1118shown. Time spent within the concurrent MAP / CVP ranges were individually regressed on the outcome AKI. The Odds ratio point estimates are displayed for each range and are color coded by the following scale (above OR=1 indicating higher odds of AKI vs less than OR=1 indicating lower odds of AKI). Darker scales are used to demonstrate the strength of OR. Ranges in gray (b) had insufficient observations for full adjustment. An asterix denotes p value< 05. The 95% CI are also displayed under the OR point estimate. A nonlinear interaction is visualized between MAP, CVP, and AKI. Odds of AKI appear lowest in the right lower comer with odds of AKI increasing towards the upper left comer.

[0075] FIG. 12 shows various data collected during studies, and in particular, a summary of association data between 5 minutes spent in concurrent MAP / CVP ranges and secondary outcome acute kidney injury 48 hours adjusted for all covariates and for multiple comparisons according to examples of the present disclosure with scales 1202, 1204, 1206, 1208, 1210, 1212, 1214, 1216, and 1218 shown. From FIG. 9, the 95%CI of each MAP / CVP are adjusted for multiple comparisons using the CMA method.

[0076] An additional 42 subjects developed AKI within 7 days for a total of 380 / 1199 (32%). Among those who developed AKI by day 7 after surgery, 312 (82%) developed stage 1 injury, 61 (16%) stage 2, and 7 (2%) stage 3. Results for the secondary outcome of AKI at 7 days were similar to the primary outcome at 48 hours for all concurrent MAP / CVP ranges (FIG. 12 and FIG. 13). Point estimates were somewhat lower for the secondary outcome perhaps reflecting the impact of additional exposures on outcome that are more remote from intraoperative perturbations in hemodynamics.

[0077] Discussion

[0078] FIG. 13 A and FIG. 13B show various data collected during studies, and in particular, a summary of association between 5 minutes spent in concurrent MAP / CVP ranges and secondary acute kidney injury outcome (FIG. 13 A) unadjusted and (FIG. 13 A) adjusted for all covariates according to examples of the present disclosure with scales 1302, 1304, 1306, 1308, 1310, 1312, 1314, 1316, 1318, and 1320 and with scales 1322, 1324, 1326, 1328, 1330, 1332, 1324, and 1336 shown. Time spent within the concurrent MAP / CVP ranges were individually regressed on the outcome AKI. The Odds ratio point estimates are displayed for each range and are color coded by the following scale (red: above OR=1 indicating higher odds of AKI vs less than OR=1 indicating lower odds of AKI). Darker scales are used to demonstrate the strength of OR. Ranges in gray (FIG. 13B) had insufficient observations for full adjustment. An asterix denotes p value< 05. The 95% CI are also displayed under the OR point estimate.A nonlinear interaction is visualized between MAP, CVP, and AKI. Odds of AKI appear lowest in the right lower comer with odds of AKI increasing towards the upper left corner.

[0079] FIG. 14 shows various data collected during studies, and in particular, a summary of association data between 5 minutes spent in concurrent MAP / CVP ranges and secondary acute kidney injury outcome adjusted for all covariates and for multiple comparisons according to examples of the present disclosure with scales 1402, 1404, 1406, 1408, 1410, 1412, 1414, and 1416 shown. From FIG. 13B, the 95%CI of each MAP / CVP are adjusted for multiple comparisons using the CMA method.

[0080] FIG. 15A and FIG. 15B show various data collected during studies, and in particular, a summary of identified 5 zones of Mean Arterial Pressure and Central Venous Pressure evaluated together with either increased risk (Zones 4 and 5) or protection (zones 1 and 2) from AKI from the data in FIG. 14 according to examples of the present disclosure with scales 1502, 1504, 1506, 1508, 1510, 1512, 1514, 1516, and 1520 shown and scales 1522, 1524, 1526, 1528, and 1530 shown.

[0081] In this retrospective cohort study of 1199 patients undergoing CAB surgery, observed were significant associations of CSA-AKI with several ranges of MAP and CVP separately and with the concurrent MAP / CVP exposure as it occurs in vivo. Consistent with previous reports, also observed were significantly increased odds of AKI associated with MAPs below 60 mmHg. Significant increase in risk of AKI associated with CVPs greater than 12 mmHg, even in normotensive subjects, were also observed, confirming and extending recent work in this area. For the first time, it is established here that a significantly reduced risk of CSA-AKI with MAP range of 90-95 mmHg separately, and for concurrent MAP / CVP ranges of MAP 65- 105 / CVP 4-8. No evidence was found to suggest that a CVP of 8-10 mmHg was associated with reduced renal injury when examining associations of CSA-AKI with CVP separately or with MAP / CVP concurrently.

[0082] Baseline characteristics of the CAB cohort were comparable to other CAB cohorts and the observed 28% incidence of AKI is in the range previously reported. As in previous studies, baseline comorbidities like older age, baseline creatinine, and chronic lung disease and intraoperative factors such as blood transfusions and length of CPB and surgery were associated with AKI.

[0083] Exposure to 5 minutes or more of intraoperative hypotensive, defined as a MAP less than 65 mmHg, was quite common in our cohort. Increased odds of AKI were observed at MAPs less than 65 mmHg, with significant associations at MAPs less than 60 mmHg,confirming multiple prior reports. Analyses examined the full set of MAP ranges and did not assume a linear relation between MAP and CSA-AKI across that range. MAP ranges of 15 mmHg increments and an association of renal failure with every 10 minutes spent within MAP range of 50- 65 mmHg has been previously reported as has exposure below MAP of 65 mmHg. Few studies have assessed the association of AKI with MAP ranges above the hypotensive threshold. Using a novel time-in-range regression approach, a nonlinear relationship between MAP and AKI outcome is observed and described herein with significant reduction in odds of AKI associated with MAP range of 90-95 mmHg. Exposures to CVP’s greater that 12 mmHg were relatively common in the cohort. This study confirms recent studies that reported increased risk of CSA-AKI associated with CVP greater than 12 mmHg. These prior studies examined threshold effects for high CVP using cut-points above 12, 16, and 20 mmHg. Unlike these previous studies, no thresholds are set or linear associations assumed between CVP exposure and outcome it the data presented herein. Using time-in-range regression, increased risk of AKI at higher CVPs is observed; however, the data methods also revealed nonlinear associations between CVP and AKI outcome, with reduced odds of AKI at lower CVPs.

[0084] FIG. 15A and FIG. 15B show various data collected during studies, and in particular, a summary of identified 5 zones of Mean Arterial Pressure and Central Venous Pressure evaluated together with either increased risk (Zones 4 and 5) or protection (Zones 1 and 2) from AKI from the data in FIG. 14 according to examples of the present disclosure.

[0085] Following from the prior series of analyses that identified 5 zones of Mean Arterial Pressure and Central Venous Pressure evaluated together with either increased risk (Zones 4 and 5) or protection (Zones 1 and 2) from AKI, we assessed the zones 4 / 5 as shown in FIG. 15A and FIG. 15B.

[0086] The series of studies and results below diagram how we have been able to build upon our results and create the clinical software for an aggregate organ (kidney) risk assessment score in the OR. Also, our results show modifications in further mean arterial and central venous pressures exposures would reduce the risk of post operative AKI in an individual patient. Thus, technology may give real time clinical decision support to reduce acute kidney injury in surgery.

[0087] FIG. 16 shows plots of the relationship between Zones 4 / 5 on AKI probability from the data in FIG. 14 according to examples of the present disclosure. First, the linearity of the relationship between Zones 4 / 5 and AKI is analyzed. It was found, as shown in FIG. 16, that it is a linear relationship, and exposure substantially increases the probability of AKI.

[0088] FIG. 17 shows plots of the relationship of Zone 3 on AKI probability from the data in FIG. 14 which has no effect on AKI probability in contrast to Zones 4 / 5 of FIG. 16 according to examples of the present disclosure.

[0089] FIG. 18 shows plots of the relationship of Zone 1 and Zone 2 exposure vs Zone 1 and Zone 2 on AKI probability from the data in FIG. 14 according to examples of the present disclosure. As shown in FIG. 18, the data shows protection in the form of reduced risk of AKI from zones 1 and 2.

[0090] A true gradient of risk between exposures is then assessed and the results from our evaluations below demonstrate the following.

[0091] Zone 4 / 5 VS Zone 3

[0092] ##

[0093] ## Simultaneous Tests for General Linear Hypotheses

[0094] ##

[0095] ## Fit: glm(formula = as.integer(bin_aki48h) ~ val_predrenf + z45 + z3 +

[0096] ## zl2, family = "binomial", data = dglm)

[0097] ##

[0098] ## Linear Hypotheses:

[0099] ## Estimate Std. Error z value Pr(>|z|)

[0100] ## 1 == 0 0.003113 0.001523 2.043 0.041 *

[0101] ## —

[0102] ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 0.1 " 1

[0103] ## (Adjusted p values reported — single-step method)

[0104] Zone 3 VS Zone 1 / 2

[0105] ##

[0106] ## Simultaneous Tests for General Linear Hypotheses

[0107] ##

[0108] ## Fit: glm(formula = as.integer(bin_aki48h) ~ val_predrenf + z45 + z3 +

[0109] ## zl2, family = "binomial", data = dglm)

[0110] ##

[0111] ## Linear Hypotheses:

[0112] ## Estimate Std. Error z value Pr(>|z|)

[0113] ## 1 == 0 0.003923 0.002266 1.731 0.0835 .

[0114] ## —

[0115] ## Signif. codes: 0 '***' 0.001 '**' 0.01 0.05 0.1 ' ' 1

[0116] ## (Adjusted p values reported — single-step method)

[0117] Zone 4 / 5 VS Zone 1 / 2

[0118] ##

[0119] ## Simultaneous Tests for General Linear Hypotheses

[0120] ##

[0121] ## Fit: glm(formula = as.integer(bin_aki48h) ~ val_predrenf + z45 + z3 +

[0122] ## zl2, family = "binomial", data = dglm)

[0123] ##

[0124] ## Linear Hypotheses:

[0125] ## Estimate Std. Error z value Pr(>|z|)

[0126] ## 1 == 0 0.007035 0.001609 4.373 1.23e-05 ***

[0127] ## —

[0128] ## Signif. codes: 0 '***' 0.001 '**' 0.01 0.05 0.1 " 1

[0129] ## (Adjusted p values reported — single-step method)

[0130] FIG. 19 shows plots of the exposure during the different time phases of surgery that contribute to AKI probability from the data in FIG. 15 according to examples of the present disclosure. Next, the exposure during the different time phases of surgery contributes to AKI is demonstrated in FIG. 20.

[0131] Then, a risk aggregation is used through these different time phases and determined trajectories of AKI. For this, the different models below can be used.• AKI ~ baseline risk factors (age, sex, . . . , predmort).• Model 1 :AKI ~ al * preCPB ztotal + bl * preCPB_proctime + offset(logit(prob_m0)). o Included prob mO as an offset to allow “preCPB ztotal” to condition on the baseline risk. From this model, extract the probability of AKI and call it prob ml, which incorporates knowledge from both baseline risk factors and the preCPB period.. Model 2:AKI ~ a2 * intraCPB ztotal + b2 * intraCPB_proctime + offset(logit(prob_ml)). o Included prob ml as an offset to allow “intraCPB ztotal” to condition on information from baseline and preCPB. From this model extract prob_m2.. Model 3:AKI ~ a3 * postCPB ztotal + b3 * postCPB_proctime + offset(logit(prob_m2)).

[0132] FIG. 20 show plots of trajectory of probability (AKI) from the data in FIG. 14 according to examples of the present disclosure.

[0133] These numbers of patients appear to have the possibility for change based on our assessment.

[0134] ##

[0135] ## Moderate Decrease Small Decrease Stable Small Increase

[0136] ## 90 248 617 101

[0137] ## Moderate Increase Large Increase

[0138] ## 98 45

[0139] Lastly, a robust risk estimator is used to develop a flexible risk model where a new type of regression is used that does not depend on any shape of the data but could determine it. FIG. 21 show plots of Harm Accumulation Function versus duration in minutes for different adjected K values for a harm accumulation model using flexible risk patterning according to examples of the present disclosure. As shown in FIG. 21, the plots use this function to plot the shape of the risk to exposure to abnormal mean arterial pressure. This is useful to then accurately develop real time organ aggregate injury calculation.

[0140] Finally, all prior results with the new regression methodology are used to write software that can incorporate in real time the mean arterial pressure data / central venous pressure data and accumulate all exposure during the surgery up until that point to accurately forecast AKI risk including how additional exposures will increase risk (Zone 4 / 5 or decrease risk Zone 1 / 2).

[0141] Merging data from Society of Thoracic Surgery (STS) Database with Electronic Health Record (EHR) data

[0142] We sought to develop the most complete and accurate database possible. We successfully merged data from several sources with time stamps but found several errors and inconsistencies in this process. For instance, the STS may not clearly delineate how to calculate total time on CPB when CPB is stopped and started. We found multiple missing time stamps for CPB start and stop. To manually validate these times we went back into the chart for any inconsistency and determined the time at which mechanical ventilation was stopped (CPB start) with loss of pulse pressure in the arterial blood pressure line which would be consistent with the start of CPB. Similarly, for any missing covariate or aberrant value, rather than exclude patients, we manually extracted missing covariates from the original record in the EHR.

[0143] Preprocessing of Hemodynamic Data for MAP and CVP

[0144] Hemodynamic data were managed as previously published with modification.1,2Measurements for mean arterial pressure (MAP), systolic blood pressure (SBP), and diastolic blood pressure (DBP) were available in the EHR from non-invasive blood pressure (NIBP) cuffs and from arterial lines at a maximum of 1 -minute resolution. Measurements for central venous pressure (CVP) were also available from the EHR at a maximum of 1 minute resolution. All time stamped measurements for blood pressure and CVP (recorded in mm Hg) were extracted from the EHR along with time-stamped intraoperative events (e.g. anesthesia start, CPB start, aortic cross clamp, etc.). We indexed pressure measurements to the timestamp of surgery start, excluding values with timestamps less than the surgery start time or greater than the surgery end time.

[0145] Values for CVP<0 during CPB were replaced with zero (artifactual values <0 are known to occur due to negative pressure exerted on IVC cannula by CPB circuit). All other out of range values were designated as missing.

[0146] Missing values at 1 -minute intervals in the MAP and CVP time series were linearly interpolated if the period of missingness was less than or equal to 15 consecutive minutes. Gaps >15 minutes were not interpolated and left as missing. The start point of the interpolation was the mean of MAP (or CVP) in the three minutes prior to the first missing value, and the end point of the interpolation was the mean of MAP (or CVP) in the three minutes after the last missing value.

[0147] Assessment of Data Quality

[0148] To evaluate data quality for MAP, we quantified the total number of minutes and the percent of time data were missing during each of the following intervals: 1) from first MAP to last MAP within anesthesia start to end; 2) from first MAP to start of CPB; 3) during CPB; and 4) from CBP end to end of surgery. To evaluate data quality for CVP, we took a similar approach with modification. Because CVP starts after central line placement, we used the timestamp of central line placement in the chart as the start time for the CVP waveform. Total time of data recordings during each surgical period and the average missingness per subject are shown below:

[0149] Summary of Missing DataTotal time from first. Minutes Minutes % Minutes % Minutes to last MAP missing Missing MissingSurgery recording MAP MAP ' CVP periodMean (SD) Mean (SD) Mean (SD) Mean (SD)Pre-CPB 156 (45) 0.38 (4.55) 54 (29) 0.25 (3.2) 6 (20)Intra-CPB 1 16 (58) 0.63 (7.77) 2 0.37 (4.4) 18 (29)Post-CPB 94 (32) 0.29 (3.00) 6.7 (16) 0.24 (2.5) 7.1

[0150] Exclusion Criteria for Insufficient Hemodynamic Data

[0151] Insufficient hemodynamic data was defined for each of following three distinct periods of surgery: (1) from the start of the first observed MAP measurement after the start of anesthesia to the start of cardiopulmonary bypass, (2) during cardiopulmonary bypass, and (3) end of cardiopulmonary bypass to last MAP measurement during anesthesia. The first and last MAP measurements were used instead of the start and end of anesthesia because intraoperative hemodynamic data recordings typically begin after the start of anesthesia (accounting for placement of non-invasive and invasive recording devices) and end before anesthesia end (accounting for transition to transport monitoring devices). Within each of these three periods, the total number of minutes with a MAP measurement was calculated. Individuals were excluded if, in any of the three time periods, there was less than 20 minutes of MAP measurements and more than 50% of minutes during that period had missing MAP values after interpolation. The same process was repeated for CVP. Only 117 of 1316 subjects required exclusion due to insufficient hemodynamic data.

[0152] Statistical Analysis Supplement: The “Fine-Mapping” Approach

[0153] MAP values were divided into 14 different ranges in increments of 5 mmHg from 45- 115 mmHg (eg 45-50, 50-55, etc.), and CVP values were divided into 10 ranges in increments of 2 mmHg from 0-20 mmHg (0-2, 2-4, etc.). For each MAP or CVP range, we counted the “time spent” as the cumulative number of minutes within each specific range. For the joint concurrent MAP / CVP exposure, MAP was divided by increments of 10 mmHg and CVP by increments of 2 mmHg. This resulted in 70 joint MAP / CVP ranges (eg. MAP 45-55 / CVP 8-10 mmHg, MAP 55-65 mm Hg / CVP 8-10 mm Hg, etc...). For each joint range, we determined the MAP and CVP values that occurred concurrently for each intraoperative minute and counted “time spent” as the cumulative number of minutes when MAP and CVP were both in range at the same time.

[0154] We ran separate regression models for each hemodynamic range, regressing AKI on total minutes in each range generating separate odds ratios / 95% CI for each range. Results arereported unadjusted and adjusted for covariates. We adopted several strategies to account for potential for Type 1 error from testing multiple ranges. First, we include an adjustment for correlation and multiplicity across the hemodynamic ranges that has been used in other highly correlated datasets. Correlation and multiplicity adjustment was developed to assess the possibility of Type I error in the setting of many comparisons that have a dependence structure. This method was developed by author Dr. Crainiceanu but has not before been used in hemodynamics research.1Second, we conducted sensitivity analyses to examine associations of AKI with a series of wider ranges for MAP and CVP. The use of wider ranges mitigates bias from correlation between adjacent ranges (wider ranges are less correlated with each other). Finally, we included all 5 zones in a single regression model (correlations across the zones were all <0.5). In this model, risk estimates for each zone are adjusted for time spent in all others.

[0155] Covariates by tertiles of minutes of exposure to MAP<65 mmHgAbbreviations: BMI, body mass index; ACE / ARB, angiotensin converting enzyme inhibitor / angiotensin II receptor blocker; STS, Society of Thoracic Surgeons Values are mean (SD) or otherwise n (%) as noted

[0156] Covariates by tertiles of minutes of exposure to CVP>12 mmHgAbbreviations: BMI, body mass index; ACE / ARB, angiotensin converting enzyme inhibitor / angiotensin II receptor blocker; STS, Society of Thoracic Surgeons Values are mean (SD) or otherwise n (%) as noted

[0157] FIG. 22 shows a plot of the summary of exposure to narrow hemodynamic ranges for mean arterial pressure (MAP) according to examples of the present disclosure. Mean cumulative time (SD) spent in each MAP range in minutes (lower bars); percent of individuals with at least 5 minutes in each range (higher bars).

[0158] FIG. 23 and FIG. 24 show plots of the sensitivity analyses for associations of CSA-AKI with a series of MAP ranges in increments of 10 and 15mm Hg, respectively, according to examples of the present disclosure. Odds ratios and 95% Cis shown from each regression model.

[0159] FIG. 25 shows a plot of the sensitivity analyses for associations of AKI with MAP ranges stratified by phase of surgery according to examples of the present disclosure. Odds ratios and 95% Cis shown for each regression model adjusted for all covariates from the primary analysis. Odds ratios not shown when data was too sparse to generate point estimates for risk.

[0160] FIG. 26 shows a plot of the summary of exposure to narrow hemodynamic ranges for central venous pressure (CVP) according to examples of the present disclosure. Mean cumulative time (SD) spent in each CVP range in minutes (lower bars); percent of individuals with at least 5 minutes in each range (higher bars).

[0161] FIG. 27 shows a plot of the sensitivity analyses for association of AKI with CVP ranges stratified by phase of surgery according to examples of the present disclosure. Odds ratios and 95% Cis shown for each regression model adjusted for all covariates from the primary analysis.

[0162] FIG. 28 and FIG. 29 shows the summary of Exposure within Joint Concurrent MAP / CVP ranges, according to examples of the present disclosure, respectively where FIG. 28 shows Mean (SD) cumulative minutes for each range and FIG. 29 shows number of subjects (%) with at least 5 minutes within range.

[0163] FIG. 30 and FIG. 31 show data of the association of acute kidney injury (AKI) with cumulative time spent in narrow ranges of joint concurrent MAP / CVP according to examplesof the present disclosure with scales 3002, 3004, 3006, 3008, 3010, 3012, 3014, 3016, and 3018 shown and scales 3102, 3104, 3106, 3108, 3110, 3112, 3114, 3116, and 3118 shown. MAP and CVP were examined concurrently at 1 -minute intervals. We determined the cumulative time when both MAP and CVP were within each of 70 joint MAP / CVP ranges. Separate logistic regression models estimated odds ratios (OR) and 95% confidence intervals for AKI within each joint MAP / CVP range. ORs are expressed per five minutes in range. ORs >=1.03 are shaded red, ORs<=0.97 are shaded, with intensity of color denoting increasing effect size. Ranges shaded in gray indicate that data were too sparse to have confidence in point estimates after adjustment for covariates. An asterisk denotes p <.05. FIG. 30 shows data for ORs / 95% Cis in each of the 70 joint MAP / CVP ranges adjusted for all covariates. FIG. 31 shows data for ORs / 95% Cis in each of the 70 joint MAP / CVP ranges adjusted for all covariates plus adjustments for correlation of data and multiple comparisons.

[0164] FIG. 32 shows computer-generated display for a patient that were generated using the methods described herein according to examples of the present disclosure. For example, the computer-generated displays can have fields that include a patient identifier, a surgery progress graph that shows the status / progress of the surgery, a zone status field that shows the current zone status of the patient, a graphic for time in each zone range, and a graphic for AKI probability versus surgery time that updates as new data is acquired.

[0165] FIG. 33 shows a flow chart 3300 for a computer-implemented method for reducing acute kidney injury during cardiac surgery. The method comprises measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP), as in 3302. For example, the MAP data range is 45-115 mmHg and the CVP data range is 0-20 mmHg, the MAP and the CVP are measured and recorded at a resolution of no greater than 1 minute resolution, and the MAP and the CVP are measured and recorded from an intraoperative electronic health record.

[0166] The method continues by storing the MAP and CVP data as a raw data series, as in 3304. The method continues by preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data, as in 3306. The method continues by assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range, as in 3308. For example, the first pressure window is 5 mmHg and the second pressure window is 2 mmHg.

[0167] The method continues by simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window, as in 3310. The method continues by determining an outcome of the cardiac surgery, as in 4812. The method continuesby processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury, as in 3314.

[0168] Herein is presented an association between reduced odds of AKI and time below CVP of 8 mmHg from both the individual CVP and concurrent MAP / CVP analyses. Specifically, every 5 minutes spent in CVP range 4-8 mmHg was associated with reduced risk of AKI for MAP values ranging from 65-105 mmHg in analyses of concurrent MAP / CVP exposure. No evidence was found to suggest that a CVP of 8-10 mmHg is associated with improved AKI outcome at any MAP range in analyses of concurrent MAP / CVP exposure or using CVP exposure alone. The findings from this observational study cannot be used to determine the optimal CVP in any individual patient; however, the 2023 clinical practice guideline that recommends targeting a CVP to 8-10 mmHg as a universal strategy to reduce risk of C SA- AKI is not supported. Lower CVPs as associated with lower risk of AKI is consistent with previous reports in patients with heart disease. For example, in a large cohort of patients with heart disease from a variety of causes, each 1 mmHg increase in CVP above 3 mmHg was associated with impaired renal function and mortality, and in a cardiac surgery cohort, lower CVP measured once at the end of surgery was associated with lower risk of AKI.

[0169] In some embodiments, any of the methods of the present disclosure may be executed by a computing system. FIG. 34 illustrates an example of such a computing system 3400, in accordance with some embodiments. The computing system 3400 may include a computer or computer system 3401A, which may be an individual computer system 3401A or an arrangement of distributed computer systems. The computer system 3401 A includes one or more analysis module(s) 3402 configured to perform various tasks according to some embodiments, such as one or more methods disclosed herein. To perform these various tasks, the analysis module 3402 executes independently, or in coordination with, one or more processors 3404, which is (or are) connected to one or more storage media 3406. The processor(s) 3404 is (or are) also connected to a network interface 3407 to allow the computer system 3401 A to communicate over a data network 3409 with one or more additional computer systems and / or computing systems, such as 340 IB, 3401C, and / or 340 ID (note that computer systems 3401B, 3401C and / or 3401D may or may not share the same architecture as computer system 3401A, and may be located in different physical locations, e.g., computer systems 3401A and 3401B may be located in a processing facility, while in communication with one or more computer systems such as 3401C and / or 340 ID that are located in one or more data centers, and / or located in varying countries on different continents). A processor can include amicroprocessor, microcontroller, processor module or subsystem, programmable integrated circuit, programmable gate array, or another control or computing device.

[0170] The storage media 3406 can be implemented as one or more computer-readable or machine-readable storage media and can store various types of data that can be accessible by operating system 3410. The storage media 3406 can be connected to or coupled with a neuromodulation machine learning module(s) 3408. Note that while in the example embodiment of FIG. 34 storage media 3406 is depicted as within computer system 3401 A, in some embodiments, storage media 3406 may be distributed within and / or across multiple internal and / or external enclosures of computing system 3401A and / or additional computing systems. Storage media 3406 may include one or more different forms of memory including semiconductor memory devices such as dynamic or static random access memories (DRAMs or SRAMs), erasable and programmable read-only memories (EPROMs), electrically erasable and programmable read-only memories (EEPROMs) and flash memories, magnetic disks such as fixed, floppy and removable disks, other magnetic media including tape, optical media such as compact disks (CDs) or digital video disks (DVDs), BLURAY® disks, or other types of optical storage, or other types of storage devices. Note that the instructions discussed above can be provided on one computer-readable or machine-readable storage medium, or alternatively, can be provided on multiple computer-readable or machine-readable storage media distributed in a large system having possibly plural nodes. Such computer-readable or machine-readable storage medium or media is (are) considered to be part of an article (or article of manufacture). An article or article of manufacture can refer to any manufactured single component or multiple components. The storage medium or media can be located either in the machine running the machine-readable instructions or located at a remote site from which machine-readable instructions can be downloaded over a network for execution.

[0171] It should be appreciated that computing system 3400 is only one example of a computing system, and that computing system 3400 may have more or fewer components than shown, may combine additional components not depicted in the example embodiment of FIG. 34, and / or computing system 3400 may have a different configuration or arrangement of the components depicted in FIG. 34. The various components shown in FIG. 34 may be implemented in hardware, software, or a combination of both hardware and software, including one or more signal processing and / or application specific integrated circuits.

[0172] Further, the steps in the processing methods described herein may be implemented by running one or more functional modules in an information processing apparatus such as generalpurpose processors or application specific chips, such as ASICs, FPGAs, PLDs, or other appropriate devices. These modules, combinations of these modules, and / or their combination with general hardware are all included within the scope of protection of the invention.

[0173] The various above-described factors, models and / or other interpretation aids may be refined in an iterative fashion; this concept is applicable to embodiments of the present methods discussed herein. This can include use of feedback loops executed on an algorithmic basis, such as at a computing device (e.g., computing system 3400, FIG. 34), and / or through manual control by a user who may make determinations regarding whether a given step, action, template, model, or set of curves has become sufficiently accurate for the evaluation of the signal(s) under consideration.

[0174] Different examples of the apparatus(es) and method(s) disclosed herein include a variety of components, features, and functionalities. It should be understood that the various examples of the apparatus(es) and method(s) disclosed herein may include any of the components, features, and functionalities of any of the other examples of the apparatus(es) and method(s) disclosed herein in any combination, and all of such possibilities are intended to be within the scope of the present disclosure. Many modifications of examples set forth herein will come to mind to one skilled in the art to which the present disclosure pertains having the benefit of the teachings presented in the foregoing descriptions and the associated drawings.

[0175] Reference herein to "one example" means that one or more feature, structure, or characteristic described in connection with the example is included in at least one implementation. The phrase "one example" in various places in the specification may or may not be referring to the same example. As used herein, a system, apparatus, structure, article, element, component, or hardware "configured to" perform a specified function is indeed capable of performing the specified function without any alteration, rather than merely having potential to perform the specified function after further modification. In other words, the system, apparatus, structure, article, element, component, or hardware "configured to" perform a specified function is specifically selected, created, implemented, utilized, programmed, and / or designed for the purpose of performing the specified function. As used herein,"configured to" denotes existing characteristics of a system, apparatus, structure, article, element, component, or hardware which enable the system, apparatus, structure, article, element, component, or hardware to perform the specified function without further modification. For purposes of this disclosure, a system, apparatus, structure, article, element, component, or hardware described as being "configured to" perform a particular function mayadditionally or alternatively be described as being "adapted to" and / or as being "operative to" perform that function.

[0176] Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the embodiments are approximations, the numerical values set forth in the specific examples are reported as precisely as possible. Any numerical value, however, inherently contains certain errors necessarily resulting from the standard deviation found in their respective testing measurements. Moreover, all ranges disclosed herein are to be understood to encompass any and all sub-ranges subsumed therein. For example, a range of "less than 10" can include any and all sub-ranges between (and including) the minimum value of zero and the maximum value of 10, that is, any and all sub-ranges having a minimum value of equal to or greater than zero and a maximum value of equal to or less than 10, e.g., 1 to 5. In certain cases, the numerical values as stated for the parameter can take on negative values. In this case, the example value of range stated as “less than 10” can assume negative values, e.g. -1, -2, -3, - 10, -20, -30, etc.

[0177] Furthermore, to the extent that the terms “including”, “includes”, “having”, “has”, “with”, or variants thereof are used in either the detailed description and the claims, such terms are intended to be inclusive in a manner similar to the term “comprising.” As used herein, the phrase “one or more of’, for example, A, B, and C means any of the following: either A, B, or C alone; or combinations of two, such as A and B, B and C, and A and C; or combinations of A, B and C.

[0178] The foregoing description, for purpose of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. Moreover, the order in which the elements of the methods are illustrated and described may be re-arranged, and / or two or more elements may occur simultaneously. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated.Conclusion

[0179] In summary, in a retrospective observational cohort of subjects undergoing CAB surgery, we found an association of CSA-AKI with intraoperative time spent in MAP ranges <60 and CVP ranges >12 mmHg. For the first time, we report a reduction in odds of CSA-AKIassociated with intraoperative CVP range of 4-6 mmHg and concurrent MAP / CVP ranges of MAP 65-105 / CVP 4-8 mmHg. Contrary to some recommendations, we found no evidence to suggest maintaining a CVP 8-10 mmHg is associated with protection from CSA-AKI.

Claims

CLAIMSWhat is claimed is:

1. A computer-implemented method for reducing acute kidney injury during cardiac surgery, comprising: measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP); storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury.

2. The method of claim 1, wherein the MAP data range is 45-115 mmHg and the CVP data range is 0-20 mmHg.

3. The method of claim 1, wherein the MAP and the CVP are measured and recorded at a resolution of no greater than 1 minute resolution.

4. The method of claim 3, wherein the MAP and the CVP are measured and recorded from an intraoperative electronic health record.

5. The method of claim 3, wherein the first pressure window is 5 mmHg and the second pressure window is 2 mmHg.

6. A computer system comprising: a hardware processor;a non-transitory computer-readable medium that stores instruction, that when executed by the hardware processor, perform a method for reducing acute kidney injury during cardiac surgery, comprising: measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP); storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury.

7. The computer system of claim 6, wherein the MAP data range is 45- 115 mmHg and the CVP data range is 0-20 mmHg.

8. The computer system of claim 6, wherein the MAP and the CVP are measured and recorded at a resolution of no greater than 1 minute resolution.

9. The computer system of claim 8, wherein the MAP and the CVP are measured and recorded from an intraoperative electronic health record.

10. The computer system of claim 8, wherein the first pressure window is 5 mmHg and the second pressure window is 2 mmHg.

11. A non-transitory computer-readable medium that stores instruction, that when executed by a hardware processor, perform a method for reducing acute kidney injury during cardiac surgery, comprising: measuring and recording mean arterial pressure (MAP) and central venous pressure (CVP);storing the MAP and CVP data as a raw data series; preprocessing the raw data series as preprocessed data, the preprocessing including one or more of managing artifacts, outliers and missing data; assessing the preprocessed data using a first pressure window across a MAP data range and a second pressure window across a CVP data range; simultaneously tracking a number of minutes within each MAP data range window and within each CVP data range window; determining an outcome of the cardiac surgery; and processing a number of minutes within MAP windows, CVP windows for outcomes to reduce kidney injury.

12. The non-transitory computer-readable medium of claim 11, wherein the MAP data range is 45-115 mmHg and the CVP data range is 0-20 mmHg.

13. The non-transitory computer-readable medium of claim 11, wherein the MAP and the CVP are measured and recorded at a resolution of no greater than 1 minute resolution.

14. The non-transitory computer-readable medium of claim 13, wherein the MAP and the CVP are measured and recorded from an intraoperative electronic health record.

15. The non-transitory computer-readable medium of claim 13, wherein the first pressure window is 5 mmHg and the second pressure window is 2 mmHg.

Images