Roger G. Mark

Multiparameter Intelligent Monitoring in Intensive Care II: A public-access intensive care unit database*

Objective:We sought to develop an intensive care unit research database applying automated techniques to aggregate high-resolution diagnostic and therapeutic data from a large, diverse population of adult intensive care unit patients. This freely available database is intended to support epidemiologic research in critical care medicine and serve as a resource to evaluate new clinical decision support and monitoring algorithms. Design:Data collection and retrospective analysis. Setting:All adult intensive care units (medical intensive care unit, surgical intensive care unit, cardiac care unit, cardiac surgery recovery unit) at a tertiary care hospital. Patients:Adult patients admitted to intensive care units between 2001 and 2007. Interventions:None. Measurements and Main Results:The Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II) database consists of 25,328 intensive care unit stays. The investigators collected detailed information about intensive care unit patient stays, including laboratory data, therapeutic intervention profiles such as vasoactive medication drip rates and ventilator settings, nursing progress notes, discharge summaries, radiology reports, provider order entry data, International Classification of Diseases, 9th Revision codes, and, for a subset of patients, high-resolution vital sign trends and waveforms. Data were automatically deidentified to comply with Health Insurance Portability and Accountability Act standards and integrated with relational database software to create electronic intensive care unit records for each patient stay. The data were made freely available in February 2010 through the Internet along with a detailed users guide and an assortment of data processing tools. The overall hospital mortality rate was 11.7%, which varied by critical care unit. The median intensive care unit length of stay was 2.2 days (interquartile range, 1.1–4.4 days). According to the primary International Classification of Diseases, 9th Revision codes, the following disease categories each comprised at least 5% of the case records: diseases of the circulatory system (39.1%); trauma (10.2%); diseases of the digestive system (9.7%); pulmonary diseases (9.0%); infectious diseases (7.0%); and neoplasms (6.8%). Conclusions:MIMIC-II documents a diverse and very large population of intensive care unit patient stays and contains comprehensive and detailed clinical data, including physiological waveforms and minute-by-minute trends for a subset of records. It establishes a new public-access resource for critical care research, supporting a diverse range of analytic studies spanning epidemiology, clinical decision-rule development, and electronic tool development.

PhysioNet: a Web-based resource for the study of physiologic signals

Free access to a signals archive and a signal processing/analysis software library fosters online collaboration. This article aims to introduce PhysioNet as a resource to the biomedical research community. After a capsule summary of its history and goals, we discuss what PhysioNet offers to researchers, describe some of the technology needed to support these functions, and conclude with observations gleaned from PhysioNets first year of service.

The MIT-BIH Arrhythmia Database on CD-ROM and software for use with it

A compact-disk ROM containing the Massachusetts Institute of Technology (MIT)-Bostons Beth Israel Hospital (BIH) Arrhythmia Database as well as a large number of supplementary recordings assembled for various research projects was produced. In all, the CD-ROM contains approximately 600 megabytes of digitized electrocardiograms (ECG) recordings, most with beat-by-beat annotations, having a total duration in excess of 200 hours. The CD-ROM format makes this substantial collection of ECGs accessible to researchers with PCs as well as those with larger computer systems. The contents of the CD-ROM and the issues involved in its production are described. Software for use with the CD-ROM as well as for development of similar databases is also described.<<ETX>>

Power spectral density of unevenly sampled data by least-square analysis: performance and application to heart rate signals

This work studies the frequency behavior of a least-square method to estimate the power spectral density of unevenly sampled signals. When the uneven sampling can be modeled as uniform sampling plus a stationary random deviation, this spectrum results in a periodic repetition of the original continuous time spectrum at the mean Nyquist frequency, with a low-pass effect affecting upper frequency bands that depends on the sampling dispersion. If the dispersion is small compared with the mean sampling period, the estimation at the base band is unbiased with practically no dispersion. When uneven sampling is modeled by a deterministic sinusoidal variation respect to the uniform sampling the obtained results are in agreement with those obtained for small random deviation. This approximation is usually well satisfied in signals like heart rate (HR) series. The theoretically predicted performance has been tested and corroborated with simulated and real HR signals. The Lomb method has been compared with the classical power spectral density (PSD) estimators that include resampling to get uniform sampling. The authors have found that the Lomb method avoids the major problem of classical methods: the low-pass effect of the resampling. Also only frequencies up to the mean Nyquist frequency should be considered (lower than 0.5 Hz if the HR is lower than 60 bpm). It is concluded that for PSD estimation of unevenly sampled signals the Lomb method is more suitable than fast Fourier transform or autoregressive estimate with linear or cubic interpolation. In extreme situations (low-HR or high-frequency components) the Lomb estimate still introduces high-frequency contamination that suggest further studies of superior performance interpolators. In the case of HR signals the authors have also marked the convenience of selecting a stationary heart rate period to carry out a heart rate variability analysis.

An open-source algorithm to detect onset of arterial blood pressure pulses

In this paper, we present an effective algorithm for detecting the onset of arterial blood pressure (ABP) pulses. The algorithm employs a windowed and weighted slope sum function (SSF) to extract ABP waveform features. Adaptive thresholding and search strategies are applied to the SSF signal to detect ABP pulses and to determine their onsets. Two evaluation procedures were employed. First, pulse detection accuracy was evaluated by comparing the algorithms pulse detections with reference ECG annotations using the MIT-BIH Polysomnographic Database. The algorithm detected 99.31% of the 368,364 beats annotated in the ECG. Second, the accuracy of pulse onset determination was established using a newly created, manually-edited reference ABP signal database. For 96.41% of the 39,848 beats in the reference database, the difference between the manually-edited and algorithm-determined ABP pulse onset was less than or equal to 20 ms. The C source code of the algorithm has been contributed to PhysioToolkit and is freely available from the PhysioNet website (http://www.physionet.org).

Risk Factors for ARDS in Patients Receiving Mechanical Ventilation for > 48 h

BACKGROUND Low tidal volume (Vt) ventilation for ARDS is a well-accepted concept. However, controversy persists regarding the optimal ventilator settings for patients without ARDS receiving mechanical ventilation. This study tested the hypothesis that ventilator settings influence the development of new ARDS. METHODS Retrospective analysis of patients from the Multi Parameter Intelligent Monitoring of Intensive Care-II project database who received mechanical ventilation for > or = 48 h between 2001 and 2005. RESULTS A total of 2,583 patients required > 48 h of ventilation. Of 789 patients who did not have ARDS at hospital admission, ARDS developed in 152 patients (19%). Univariate analysis revealed high peak inspiratory pressure (odds ratio [OR], 1.53 per SD; 95% confidence interval [CI], 1.28 to 1.84), increasing positive end-expiratory pressure (OR, 1.35 per SD; 95% CI, 1.15 to 1.58), and Vt (OR, 1.36 per SD; 95% CI, 1.12 to 1.64) to be significant risk factors. Major nonventilator risk factors for ARDS included sepsis, low pH, elevated lactate, low albumin, transfusion of packed RBCs, transfusion of plasma, high net fluid balance, and low respiratory compliance. Multivariable logistic regression showed that peak pressure (OR, 1.31 per SD; 95% CI, 1.08 to 1.59), high net fluid balance (OR, 1.3 per SD; 95% CI, 1.09 to 1.56), transfusion of plasma (OR, 1.26 per SD; 95% CI, 1.07 to 1.49), sepsis (OR, 1.57; 95% CI, 1.00 to 2.45), and Vt (OR, 1.29 per SD; 95% CI, 1.02 to 1.52) were significantly associated with the development of ARDS. CONCLUSIONS The associations between the development of ARDS and clinical interventions, including high airway pressures, high Vt, positive fluid balance, and transfusion of blood products, suggests that ARDS may be a preventable complication in some cases.

Long-term ST database: a reference for the development and evaluation of automated ischaemia detectors and for the study of the dynamics of myocardial ischaemia

The long-term ST database is the result of a multinational research effort. The goal was to develop a challenging and realistic research resource for development and evaluation of automated systems to detect transient ST segment changes in electrocardiograms and for supporting basic research into the mechanisms and dynamics of transient myocardial ischaemia. Twenty-four hour ambulatory ECG records were selected from routine clinical practice settings in the USA and Europe, between 1994 and 2000, on the basic of occurrence of ischaemic and non-ischaemic ST segment changes. Human expert annotators used newly developed annotation protocols and a specially developed interactive graphic editor tool (Semia) that supported paperless editing of annotations and facilitated international co-operation via the Internet. The database contains 86 two- and three-channel 24h annotated ambulatory records from 80 patients and is stored on DVD-ROMs. The database annotation files contain ST segment annotations of transient ischaemic (1155) and heart-rate related ST episodes and annotations of non-ischaemic ST segment events related to postural changes and conduction abnormalities. The database is intended to complement the European Society of Cardiology ST-T database and the MIT-BIH and AHA arrhythmia databases. It provides a comprehensive representation of ‘real-world’ data, with numerous examples of transient ischaemic and non-ischaemic ST segment changes, arrhythmias, conduction abnormalities, axis shifts, noise and artifacts.

Proton-pump inhibitor use is associated with low serum magnesium concentrations.

Although case reports link proton-pump inhibitor (PPI) use and hypomagnesemia, no large-scale studies have been conducted. Here we examined the serum magnesium concentration and the likelihood of hypomagnesemia (<1.6 mg/dl) with a history of PPI or histamine-2 receptor antagonist used to reduce gastric acid, or use of neither among 11,490 consecutive adult admissions to an intensive care unit of a tertiary medical center. Of these, 2632 patients reported PPI use prior to admission, while 657 patients were using a histamine-2 receptor antagonist. PPI use was associated with 0.012 mg/dl lower adjusted serum magnesium concentration compared to users of no acid-suppressive medications, but this effect was restricted to those patients taking diuretics. Among the 3286 patients concurrently on diuretics, PPI use was associated with a significant increase of hypomagnesemia (odds ratio 1.54) and 0.028 mg/dl lower serum magnesium concentration. Among those not using diuretics, PPI use was not associated with serum magnesium levels. Histamine-2 receptor antagonist use was not significantly associated with magnesium concentration without or with diuretic use. The use of PPI was not associated with serum phosphate concentration regardless of diuretic use. Thus, we verify case reports of the association between PPI use and hypomagnesemia in those concurrently taking diuretics. Hence, serum magnesium concentrations should be followed in susceptible individuals on chronic PPI therapy.

Outcome of critically ill patients with acute kidney injury using the Acute Kidney Injury Network criteria.

Objective:Acute kidney injury affects 5% to 7% of all hospitalized patients with a much higher incidence in the critically ill. The Acute Kidney Injury Network proposed a definition in which serum creatinine rises (>0.3 mg/dL) and/or oliguria (<0.5 mL/kg/hr) for a period of 6 hrs are used to detect acute kidney injury. Accurate urine output measurements as well as serum creatinine values from our database were used to detect patients with acute kidney injury and calculate their corresponding mortality risk and length of stay. Design:Retrospective cohort study. Setting:Seven intensive care units at a large, academic, tertiary medical center. Patients:Adult patients without evidence of end-stage renal disease with more than two creatinine measurements and at least a 6-hr urine output recording who were admitted to the intensive care unit between 2001 and 2007. Interventions:Medical records of all the patients were reviewed. Demographic information, laboratory results, charted data, discharge diagnoses, physiological data, and patient outcomes were extracted from the Multiparameter Intelligent Monitoring in Intensive Care II database using a SQL query. Measurements and Main Results:From 19,677 adult patient records, 14,524 patients met the inclusion criteria. Fifty-seven percent developed acute kidney injury during their intensive care unit stay. Inhospital mortality rates were: 13.9%, 16.4%, 33.8% for acute kidney injury 1, 2, and 3, respectively, compared with only 6.2% in patients without acute kidney injury (p < .0001). After adjusting for multiple covariates, acute kidney injury was associated with increased hospital mortality (odds ratio 1.4 and 1.3 for acute kidney injury 1 and acute kidney injury 2 and 2.5 for acute kidney injury 3; p < .0001). Using multivariate logistic regression, we found that in patients who developed acute kidney injury, urine output alone was a better mortality predictor than creatinine alone or the combination of both. Conclusions:More than 50% of our critically ill patients developed some stage of acute kidney injury resulting in a stagewise increased mortality risk. However, the mortality risk associated with acute kidney injury stages 1 and 2 does not differ significantly. In light of these findings, re-evaluation of the Acute Kidney Injury Network staging criteria should be considered.

Reduction of false arterial blood pressure alarms using signal quality assessment and relationships between the electrocardiogram and arterial blood pressure

The paper presents an algorithm for reducing false alarms related to changes in arterial blood pressure (ABP) in intensive care unit (ICU) monitoring. The algorithm assesses the ABP signal quality, analyses the relationship between the electrocardiogram and ABP using a fuzzy logic approach and post-processes (accepts or rejects) ABP alarms produced by a commerical monitor. The algorithm was developed and evaluated using unrelated sets of data from the MIMIC database. By rejecting 98.2% (159 of 162) of the false ABP alarms produced by the monitor using the test set of data, the algorithm was able to reduce the false ABP alarm rate from 26.8% to 0.5% of ABP alarms, while accepting 99.8% (441 of 442) of true ABP alarms. The results show that the algorithm is effective and practical, and its use in future patient monitoring systems is feasible.