Time trends in prostate cancer screening in Swiss primary care (2010 to 2017) - A retrospective study.

INTRODUCTION: Following years of controversy regarding screening for prostate cancer using prostate-specific antigen, evidence evolves towards a more restrained and preference-based use. This study reports the impact of landmark trials and updated recommendations on the incidence rate of prostate cancer screening by Swiss general practitioners. METHODS: We performed a retrospective analysis of primary care data, separated in 3 time periods based on dates of publications of important prostate-specific antigen screening recommendations. 1: 2010-mid 2012 including 2 updates; 2: mid 2012-mid 2014 including a Smarter Medicine recommendation; 3: mid-2014-mid-2017 maintenance period. Period 2 including the Smarter Medicine recommendation was defined as reference period. We further assessed the influence of patient's age and the number of prostate-specific-antigen (PSA) tests, by the patient and within each time period, on the mean PSA concentration. Uni- and multivariable analyses were used as needed. RESULTS: 36,800 men aged 55 to 75 years were included. 14.6% had ≥ 2 chronic conditions, 11.7% had ≥ 1 prostate-specific antigen test, (mean 2.60 ng/ml [SD 12.3]). 113,921 patient-years were covered. Data derived from 221 general practitioners, 33.5% of GP were women, mean age was 49.4 years (SD 10.0), 67.9% used prostate-specific antigen testing. Adjusted incidence rate-ratio (95%-CI) dropped significantly over time periods: Reference Period 2: incidence rate-ratio 1.00; Period 1: incidence rate-ratio 1.74 (1.59-1.90); Period 3: incidence rate-ratio 0.61 (0.56-0.67). A higher number of chronic conditions and a patient age between 60-69 years were significantly associated with higher screening rate. Increasing numbers of PSA testing per patient, as well as increasing age, were independently and significantly associated with an increase in the PSA value. CONCLUSION: Swiss general practitioners adapted screening behavior as early as evidence of a limited health benefit evolved, while using a risk-adapted approach whenever performing multiple testing. Updated recommendations might have helped to maintain this decrease. Further recommendations and campaigns should aimed at older patients with multimorbidity, to sustain a further decline in prostate-specific antigen screening practices.

Integrative hospital treatment in older patients to benchmark and improve outcome and length of stay - the In-HospiTOOL study.

BACKGROUND: A comprehensive in-hospital patient management with reasonable and economic resource allocation is arguably the major challenge of health-care systems worldwide, especially in elderly, frail, and polymorbid patients. The need for patient management tools to improve the transition process and allocation of health care resources in routine clinical care particularly for the inpatient setting is obvious. To address these issues, a large prospective trial is warranted. METHODS: The "Integrative Hospital Treatment in Older patients to benchmark and improve Outcome and Length of stay" (In-HospiTOOL) study is an investigator-initiated, multicenter effectiveness trial to compare the effects of a novel in-hospital management tool on length of hospital stay, readmission rate, quality of care, and other clinical outcomes using a time-series model. The study aims to include approximately 35`000 polymorbid medical patients over an 18-month period, divided in an observation, implementation, and intervention phase. Detailed data on treatment and outcome of polymorbid medical patients during the in-hospital stay and after 30 days will be gathered to investigate differences in resource use, inter-professional collaborations and to establish representative benchmarking data to promote measurement and display of quality of care data across seven Swiss hospitals. The trial will inform whether the "In-HospiTOOL" optimizes inter-professional collaboration and thereby reduces length of hospital stay without harming subjective and objective patient-oriented outcome markers. DISCUSSION: Many of the current quality-mirroring tools do not reflect the real need and use of resources, especially in polymorbid and elderly patients. In addition, a validated tool for optimization of patient transition and discharge processes is still missing. The proposed multicenter effectiveness trial has potential to improve interprofessional collaboration and optimizes resource allocation from hospital admission to discharge. The results will enable inter-hospital comparison of transition processes and accomplish a benchmarking for inpatient care quality.

Determinants of aggregate length of hospital stay in the last year of life in Switzerland.

BACKGROUND: In contrast to individual preferences, most people in developed countries die in health care institutions, with a considerable impact on health care resource use and costs. However, evidence about determinants of aggregate length of hospital stay in the last year preceding death is scant. METHODS: Nationwide individual patient data from Swiss hospital discharge statistics were linked with census and mortality records from the Swiss National Cohort. We explored determinants of aggregate length of hospital stay in the last year of life in N = 35,598 inpatients ≥65 years who deceased in 2007 or 2008. RESULTS: The average aggregate length of hospital stay in the last year of life was substantially longer in the German speaking region compared to the French (IRR 1.36 [95 % CI 1.32-1.40]) and Italian (IRR 1.22 [95 % CI 1.16-1.29]) speaking region of the country. Increasing age, female sex, multimorbidity, being divorced, foreign nationality, and high educational level prolonged, whereas home ownership shortened the aggregate length of hospital stay. Individuals with complementary private health insurance plans had longer stays than those with compulsory health insurance plans (IRR 1.04 [95 % CI 1.01-1.07]). CONCLUSIONS: The aggregate length of hospital stay during the last year of life was substantially determined by regional and socio-demographic characteristics, and only partially explained by differential health conditions. Therefore, more detailed studies need to evaluate, whether these differences are based on patients' health care needs and preferences, or whether they are supply-driven.

General Practitioners' vitamin K antagonist monitoring is associated with better blood pressure control in patients with hypertension--a cross-sectional database study.

BACKGROUND: Patients requiring anticoagulation suffer from comorbidities such as hypertension. On the occasion of INR monitoring, general practitioners (GPs) have the opportunity to control for blood pressure (BP). We aimed to evaluate the impact of Vitamin-K Antagonist (VKA) monitoring by GPs on BP control in patients with hypertension. METHODS: We cross-sectionally analyzed the database of the Swiss Family Medicine ICPC Research using Electronic Medical Records (FIRE) of 60 general practices in a primary care setting in Switzerland. This database includes 113,335 patients who visited their GP between 2009 and 2013. We identified patients with hypertension based on antihypertensive medication prescribed for ≥ 6 months. We compared patients with VKA for ≥ 3 months and patients without such treatment regarding BP control. We adjusted for age, sex, observation period, number of consultations and comorbidity. RESULTS: We identified 4,412 patients with hypertension and blood pressure recordings in the FIRE database. Among these, 569 (12.9%) were on Phenprocoumon (VKA) and 3,843 (87.1%) had no anticoagulation. Mean systolic and diastolic BP was significantly lower in the VKA group (130.6 ± 14.9 vs 139.8 ± 15.8 and 76.6 ± 7.9 vs 81.3 ± 9.3 mm Hg) (p < 0.001 for both). The difference remained after adjusting for possible confounders. Systolic and diastolic BP were significantly lower in the VKA group, reaching a mean difference of -8.4 mm Hg (95% CI -9.8 to -7.0 mm Hg) and -1.5 mm Hg (95% CI -2.3 to -0.7 mm Hg), respectively (p < 0.001 for both). CONCLUSIONS: In a large sample of hypertensive patients in Switzerland, VKA treatment was independently associated with better systolic and diastolic BP control. The observed effect could be due to better compliance with antihypertensive medication in patients treated with VKA. Therefore, we conclude to be aware of this possible benefit especially in patients with lower expected compliance and with multimorbidity.

Prevalence of chronic medical conditions in Switzerland: exploring estimates validity by comparing complementary data sources.

BACKGROUND: Prevalence estimates of chronic medical conditions and their multiples (multimorbidity) in the general population are scarce and often rather speculative in Switzerland. Using complementary data sources, we assessed estimates validity of population-based prevalence rates of four common chronic medical conditions with high impact on cardiovascular health (diabetes mellitus, hypertension, dyslipidemia, obesity). METHODS: We restricted our analyses to patients 15-94 years old living in the German speaking part of Switzerland. Data sources were: Swiss Health Survey (SHS, 2007, n = 13,580); Family Medicine ICPC Research using Electronic Medical Record Database (FIRE, 2010-12, n = 99,441); and hospital discharge statistics (MEDSTAT, 2009-10, n = 883,936). We defined chronic medical conditions based on use of drugs, diagnoses, and measurements. RESULTS: After a careful harmonization of the definitions, a high degree of concordance, especially regarding the age- and gender-specific distribution patterns, was found for diabetes mellitus (defined as drug use or diagnosis in SHS, drug use or diagnosis or blood glucose measurement in FIRE, and ICD-10 codes E10-14 as secondary diagnosis in MEDSTAT) and for hypertension (defined as drug use alone in SHS and FIRE, and ICD-10 codes I10-15 or I67.4 as secondary diagnosis in MEDSTAT). A lesser degree of concordance was found for dyslipidemia (defined as drug use alone in SHS and FIRE, and ICD-10 code E78 in MEDSTAT), and for obesity (defined as BMI ≥ 30 kg/m(2) derived from self-reported height and weight in SHS, from measured height and weight or diagnosis of obesity in FIRE, and ICD-10 code E66 as secondary diagnosis in MEDSTAT). MEDSTAT performed well for clearly defined diagnoses (diabetes, hypertension), but underrepresented systematically more symptomatic conditions (dyslipidemia, obesity). CONCLUSION: Complementary data sources can provide different prevalence estimates of chronic medical conditions in the general population. However, common age and sex patterns indicate that a careful harmonization of the definition of each chronic medical condition permits a high degree of concordance.

Moving to and dying in a nursing home depends not only on health - an analysis of socio-demographic determinants of place of death in Switzerland.

BACKGROUND: In developed countries generally about 7 out of 10 deaths occur in institutions such as acute care hospitals or nursing homes. However, less is known about the influence of non-medical determinants of place of death. This study examines the influence of socio-demographic and regional factors on place of death in Switzerland. DATA AND METHODS: We linked individual data from hospitals and nursing homes with census and mortality records of the Swiss general population. We differentiated between those who died in a hospital after a length of stay ≤2 days or ≥3 days, those who died in nursing homes, and those who died at home. In gender-specific multinomial logistic regression models we analysed N = 85,129 individuals, born before 1942 (i.e., ≥65 years old) and deceased in 2007 or 2008. RESULTS: Almost 70% of all men and 80% of all women died in a hospital or nursing home. Regional density of nursing home beds, being single, divorced or widowed, or living in a single-person household were predictive of death in an institution, especially among women. Conversely, homeownership, high educational level and having children were associated with dying at home. CONCLUSION: Place of death substantially depends on socio-demographic determinants such as household characteristics and living conditions as well as on regional factors. Individuals with a lower socio-economic position, living alone or having no children are more prone to die in a nursing home. Health policy should empower these vulnerable groups to choose their place of death in accordance to needs and wishes.

Therapeutic conflicts in emergency department patients with multimorbidity: a cross-sectional study.

BACKGROUND: Patients with multimorbidity are an increasing concern in healthcare. Clinical practice guidelines, however, do not take into account potential therapeutic conflicts caused by co-occurring medical conditions. This makes therapeutic decisions complex, especially in emergency situations. OBJECTIVE: The aim of this study was to identify and quantify therapeutic conflicts in emergency department patients with multimorbidity. METHODS: We reviewed electronic records of all patients ≥18 years with two or more concurrent active medical conditions, admitted from the emergency department to the hospital ward of the University Hospital Zurich in January 2009. We cross-tabulated all active diagnoses with treatments recommended by guidelines for each diagnosis. Then, we identified potential therapeutic conflicts and classified them as either major or minor conflicts according to their clinical significance. RESULTS: 166 emergency inpatients with multimorbidity were included. The mean number of active diagnoses per patient was 6.6 (SD±3.4). We identified a total of 239 therapeutic conflicts in 49% of the of the study population. In 29% of the study population major therapeutic conflicts, in 41% of the patients minor therapeutic conflicts occurred. CONCLUSIONS: Therapeutic conflicts are common among multimorbid patients, with one out of two experiencing minor, and one out of three experiencing major therapeutic conflicts. Clinical practice guidelines need to address frequent therapeutic conflicts in patients with co-morbid medical conditions.

Patient- and physician-related risk factors for hyperkalaemia in potassium-increasing drug-drug interactions.

PURPOSE: Hyperkalaemia due to potassium-increasing drug-drug interactions (DDIs) is a clinically important adverse drug event. The purpose of this study was to identify patient- and physician-related risk factors for the development of hyperkalaemia. METHODS: The risk for adult patients hospitalised in the University Hospital Zurich between 1 December 2009 and 31 December 2011 of developing hyperkalaemia was correlated with patient characteristics, number, type and duration of potassium-increasing DDIs and frequency of serum potassium monitoring. RESULTS: The 76,467 patients included in this study were prescribed 8,413 potentially severe potassium-increasing DDIs. Patient-related characteristics associated with the development of hyperkalaemia were pulmonary allograft [relative risk (RR) 5.1; p < 0.0001), impaired renal function (RR 2.7; p < 0.0001), diabetes mellitus (RR 1.6; p = 0.002) and female gender (RR 1.5; p = 0.007). Risk factors associated with medication were number of concurrently administered potassium-increasing drugs (RR 3.3 per additional drug; p < 0.0001) and longer duration of the DDI (RR 4.9 for duration ≥6 days; p < 0.0001). Physician-related factors associated with the development of hyperkalaemia were undetermined or elevated serum potassium level before treatment initiation (RR 2.2; p < 0.001) and infrequent monitoring of serum potassium during a DDI (interval >48 h: RR 1.6; p < 0.01). CONCLUSION: Strategies for reducing the risk of hyperkalaemia during potassium-increasing DDIs should consider both patient- and physician-related risk factors.

Non-urgent encounters in a Swiss medical emergency unit.

BACKGROUND: Emergency department crowding is a growing international problem. One possible reason for crowding might be the rising number of "walk-in" patients presenting with "non-urgent" health complaints. METHODS: In a retrospective cohort study in adult medical patients presenting to the emergency unit of the University Hospital Zurich, we determined the frequency of "non-urgent" encounters, examined patient characteristics predictive for such encounters, and explored the impact of a simple, non-validated triage tool on diverting "non-urgent" cases to alternate sites of primary care. RESULTS: We included 1,175 and 1,448 medical encounters before (1-31 January 2008) and after (1-31 January 2009) the implementation of the triage tool. Almost one out of three patients presented with a minor "non-urgent" health complaint (29.9% [95%CI 28.1%-31.6%]). The most common were "cough/sneezing" (7.82% [95%CI 6.79%-8.84%]), "follow-up" (6.44% [95%CI 5.50%-7.38%]), and "weakness/tiredness" (3.47% [95%CI 2.77%-4.17%]). Significant predictors for "non-urgent" encounters were young age (mean adjusted odds ratio 0.93 [95%CI 0.88-0.97] for each additional decade of life), and non-Swiss origin (adjusted odds ratio 1.18 [95%CI 1.02-1.31]). The triage tool did not divert "non-urgent" cases from the emergency unit to outpatient care (adjusted odds ratio 0.94 [95%CI 0.80-1.12]). CONCLUSION: In the emergency unit of the University Hospital Zurich, the prevalence of "non-urgent" medical encounters was substantial with one out three patients presenting with minor health complaints. Young age and non-Swiss origin were associated with increased use of the emergency unit for "non-urgent" conditions. A simple triage tool did not effectively divert "non-urgent" cases to alternates sites of primary care.

Age- and gender-related prevalence of multimorbidity in primary care: the Swiss FIRE project.

BACKGROUND: General practitioners often care for patients with several concurrent chronic medical conditions (multimorbidity). Recent data suggest that multimorbidity might be observed more often than isolated diseases in primary care. We explored the age- and gender-related prevalence of multimorbidity and compared these estimates to the prevalence estimates of other common specific diseases found in Swiss primary care. METHODS: We analyzed data from the Swiss FIRE (Family Medicine ICPC Research using Electronic Medical Record) project database, representing a total of 509,656 primary care encounters in 98,152 adult patients between January 1, 2009 and July 31, 2011. For each encounter, medical problems were encoded using the second version of the International Classification of primary Care (ICPC-2). We defined chronic health conditions using 147 pre-specified ICPC-2 codes and defined multimorbidity as 1) two or more chronic health conditions from different ICPC-2 rubrics, 2) two or more chronic health conditions from different ICPC-2 chapters, and 3) two or more medical specialties involved in patient care. We compared the prevalence estimates of multimorbidity defined by the three methodologies with the prevalence estimates of common diseases encountered in primary care. RESULTS: Overall, the prevalence estimates of multimorbidity were similar for the three different definitions (15% [95%CI 11-18%], 13% [95%CI 10-16%], and 14% [95%CI 11-17%], respectively), and were higher than the prevalence estimates of any specific chronic health condition (hypertension, uncomplicated 9% [95%CI 7-11%], back syndrome with and without radiating pain 6% [95%CI 5-7%], non-insulin dependent diabetes mellitus 3% [95%CI 3-4%]), and degenerative joint disease 3% [95%CI 2%-4%]). The prevalence estimates of multimorbidity rose more than 20-fold with age, from 2% (95%CI 1-2%) in those aged 20-29 years, to 38% (95%CI 31-44%) in those aged 80 or more years. The prevalence estimates of multimorbidity were similar for men and women (15% vs. 14%, p=0.288). CONCLUSIONS: In primary care, prevalence estimates of multimorbidity are higher than those of isolated diseases. Among the elderly, more than one out of three patients suffer from multimorbidity. Management of multimorbidity is a principal concern in this vulnerable patient population.

Clinical decision support for monitoring drug-drug-interactions and potassium-increasing drug combinations: need for specific alerts.

Computer-triggered reminders alerting physicians on every potentially harmful drug-drug-interaction (DDI) induce alert fatigue due to frequent messages of limited clinical relevance. On demand DDI-checks, however, are not commonly used by physicians. Optimal strategies for sustained quality assurance have to consider patients' risk factors and focus on the most significant DDIs only. An approach is proposed based on the analysis of concurrent prescription of potassium-sparing diuretics and potassium supplements (CPPP), which are the most frequent DDIs classified as contraindicated. Although the frequency of monitoring potassium serum levels declined during prolonged periods of CPPP, the likelihood of observing a hyperkalaemia increased. The median treatment period of CPPP was 3.3 days, whereas hyperkalaemia occurred after a median observation time of 4.5 days of CPPP. Thus, computer-triggered reminders for ordering potassium serum levels may be indicated if monitoring has been discontinued after 48h of CPPP.

Prevalence of multimorbidity in medical inpatients.

OBJECTIVE: To validate the estimates of the prevalence of multimorbidity based on administrative hospital discharge data, with medical records and chart reviews as benchmarks. DESIGN: Retrospective cohort study. SETTING: Medical division of a tertiary care teaching hospital. PARTICIPANTS: A total of 170 medical inpatients admitted from the emergency unit in January 2009. MAIN MEASURES: The prevalence of multimorbidity for three different definitions (≥2 diagnoses, ≥2 diagnoses from different ICD-10 chapters, and ≥2 medical conditions as defined by Charlson/Deyo) and three different data sources (administrative data, chart reviews, and medical records). RESULTS: The prevalence of multimorbidity in medical inpatients derived from administrative data, chart reviews and medical records was very high and concurred for the different definitions of multimorbidity (≥2 diagnoses: 96.5%, 95.3%, and 92.9% [p = 0.32], ≥2 diagnoses from different ICD-10 chapters: 86.5%, 90.0%, and 85.9% [p = 0.46], and ≥2 medical conditions as defined by Charlson/Deyo: 48.2%, 50.0%, and 46.5% [p = 0.81]). The agreement of rating of multimorbidity for administrative data and chart reviews and administrative data and medical records was 94.1% and 93.0% (kappa statistics 0.47) for ≥2 diagnoses; 86.0% and 86.5% (kappa statistics 0.52) for ≥2 diagnoses from different ICD-10 chapters; and 82.9% and 85.3% (kappa statistics 0.69) for ≥2 medical conditions as defined by Charlson/Deyo. CONCLUSION: Estimates of the prevalence of multimorbidity in medical inpatients based on administrative data, chart reviews and medical records were very high and congruent for the different definitions of multimorbidity. Agreement for rating multimorbidity based on the different data sources was moderate to good. Administrative hospital discharge data are a valid source for exploring the burden of multimorbidity in hospital settings.

Identification of a novel loss-of-function calcium channel gene mutation in short QT syndrome (SQTS6).

AIMS: Short QT syndrome (SQTS) is a genetically determined ion-channel disorder, which may cause malignant tachyarrhythmias and sudden cardiac death. Thus far, mutations in five different genes encoding potassium and calcium channel subunits have been reported. We present, for the first time, a novel loss-of-function mutation coding for an L-type calcium channel subunit. METHODS AND RESULTS: The electrocardiogram of the affected member of a single family revealed a QT interval of 317 ms (QTc 329 ms) with tall, narrow, and symmetrical T-waves. Invasive electrophysiological testing showed short ventricular refractory periods and increased vulnerability to induce ventricular fibrillation. DNA screening of the patient identified no mutation in previously known SQTS genes; however, a new variant at a heterozygous state was identified in the CACNA2D1 gene (nucleotide c.2264G > C; amino acid p.Ser755Thr), coding for the Ca(v)α(2)δ-1 subunit of the L-type calcium channel. The pathogenic role of the p.Ser755Thr variant of the CACNA2D1 gene was analysed by using co-expression of the two other L-type calcium channel subunits, Ca(v)1.2α1 and Ca(v)ß(2b), in HEK-293 cells. Barium currents (I(Ba)) were recorded in these cells under voltage-clamp conditions using the whole-cell configuration. Co-expression of the p.Ser755Thr Ca(v)α(2)δ-1 subunit strongly reduced the I(Ba) by more than 70% when compared with the co-expression of the wild-type (WT) variant. Protein expression of the three subunits was verified by performing western blots of total lysates and cell membrane fractions of HEK-293 cells. The p.Ser755Thr variant of the Ca(v)α(2)δ-1 subunit was expressed at a similar level compared with the WT subunit in both fractions. Since the mutant Ca(v)α(2)δ-1 subunit did not modify the expression of the pore-forming subunit of the L-type calcium channel, Ca(v)1.2α1, it suggests that single channel biophysical properties of the L-type channel are altered by this variant. CONCLUSION: In the present study, we report the first pathogenic mutation in the CACNA2D1 gene in humans, which causes a new variant of SQTS. It remains to be determined whether mutations in this gene lead to other manifestations of the J-wave syndrome.

Triple rule-out CT in patients with suspicion of acute pulmonary embolism: findings and accuracy.

RATIONALE AND OBJECTIVES: The aim of this study was to prospectively investigate the diagnostic value of triple rule-out computed tomography (CT) in patients suspected of having acute pulmonary embolism (PE). MATERIALS AND METHODS: A total of 125 patients with suspicion of PE, of whom 14 patients had the additional clinical suspicion of acute aortic syndrome, underwent electrocardiogram-gated triple rule-out dual-source CT. The contrast media application protocol was adjusted to obtain a homogenous attenuation of the pulmonary arteries, thoracic aorta, and coronary arteries. The diagnostic performance of triple rule-out CT was assessed by using adjudicated discharge diagnoses as reference standards. RESULTS: A total of 161 adjudicated cardiovascular discharge diagnoses were made in the 125 patients (including all true-positive and true-negative findings): acute PE was found in 26 (21%) and was excluded by CT in 99 (79%), coronary artery disease was found in 3 (3%) and was excluded by catheter angiography in 9 (6%), left ventricular systolic dysfunction was found in 2 (2%) and was excluded by echocardiography in 8 (6%), and acute aortic syndrome was found in 5 (4%) and was excluded by CT in 9 (7%) patients. Nonvascular chest disease was found in 34 (27%) and included pneumonia (n = 17), neoplasms (n = 5), fractures/osteolysis (n = 3), pericarditis (n = 2), and post-pneumonectomy syndrome (n = 1). Triple rule-out CT was normal in 53 (42%) patients. Overall sensitivity, specificity, and positive and negative predictive value of triple rule-out CT for cardiovascular disease were 100% (95% confidence interval [CI] 90-100%), 98% (95%CI 94-100%), 95% (95%CI 82-99%), and 100% (95%CI 97-100%, respectively). CONCLUSIONS: Triple rule-out CT is feasible in patients with suspicion of PE, reveals a wide range of vascular and non-vascular chest disease, and offers an excellent overall diagnostic performance.

Monitoring carbon dioxide tension and arterial oxygen saturation by a single earlobe sensor in patients with critical illness or sleep apnea.

OBJECTIVES: The purpose of the study was to evaluate a novel, combined sensor for transcutaneous monitoring of arterial oxygen saturation and carbon dioxide tension. DESIGN: The new monitoring technique was compared to established reference methods. SETTING: ICU and sleep laboratory of a university hospital. PATIENTS: Eighteen critically ill adult patients with acute respiratory failure or heart failure, and 12 patients with sleep apnea (mean [+/- SD] apnea/hypopnea index, 43 +/- 24 events per hour). MEASUREMENTS: Continuous measurements were performed over several hours by the novel heated (temperature, 42 degrees C) earlobe sensor (TOSCA; Linde Medical Sensors; Basel, Switzerland), incorporating electrochemical and optical elements for carbon dioxide measurement (PtcCO2) and pulse oximetry (SpO2), respectively. The data were compared to the results of repeated arterial blood gas analyses in critically ill patients and to simultaneous nocturnal pulse oximetry performed with different devices with earlobe or finger sensors in sleep apnea patients. RESULTS: In critically ill patients, the mean difference and limits of agreement (bias +/- 2 SDs) of transcutaneous PtcCO2 vs arterial PaCO2 were 3 +/- 7 mm Hg; the corresponding values for changes in PtcCO2 vs PaCO2 were 1 +/- 6 mm Hg. The bias +/- 2 SDs for pulse oximetric SpO2 vs arterial oxygen saturation (SaO2) were 1 +/- 4%. In sleep apnea patients, the combined earlobe sensor identified more transient oxygen desaturations, and the rate of change in oxygen saturation during events was greater compared to those with other tested pulse oximeters, indicating a faster response. CONCLUSIONS: Due to its ability to accurately assess both ventilation and oxygenation by a single transcutaneous sensor, the described noninvasive monitoring technique is a valuable tool for respiratory monitoring with potential applications in critical care and sleep medicine.

Dynamic microsimulation to model multiple outcomes in cohorts of critically ill patients.

BACKGROUND: Existing intensive care unit (ICU) prediction tools forecast single outcomes, (e.g., risk of death) and do not provide information on timing. OBJECTIVE: To build a model that predicts the temporal patterns of multiple outcomes, such as survival, organ dysfunction, and ICU length of stay, from the profile of organ dysfunction observed on admission. DESIGN: Dynamic microsimulation of a cohort of ICU patients. SETTING: 49Forty-nine ICUs in 11 countries. PATIENTS: One thousand four hundred and forty-nine patients admitted to the ICU in May 1995. INTERVENTIONS: None. MODEL CONSTRUCTION: We developed the model on all patients (n=989) from 37 randomly-selected ICUs using daily Sequential Organ Function Assessment (SOFA) scores. We validated the model on all patients (n=460) from the remaining 12 ICUs, comparing predicted-to-actual ICU mortality, SOFA scores, and ICU length of stay (LOS). MAIN RESULTS: In the validation cohort, the predicted and actual mortality were 20.1% (95%CI: 16.2%-24.0%) and 19.9% at 30 days. The predicted and actual mean ICU LOS were 7.7 (7.0-8.3) and 8.1 (7.4-8.8) days, leading to a 5.5% underestimation of total ICU bed-days. The predicted and actual cumulative SOFA scores per patient were 45.2 (39.8-50.6) and 48.2 (41.6-54.8). Predicted and actual mean daily SOFA scores were close (5.1 vs 5.5, P=0.32). Several organ-organ interactions were significant. Cardiovascular dysfunction was most, and neurological dysfunction was least, linked to scores in other organ systems. CONCLUSIONS: Dynamic microsimulation can predict the time course of multiple short-term outcomes in cohorts of critical illness from the profile of organ dysfunction observed on admission. Such a technique may prove practical as a prediction tool that evaluates ICU performance on additional dimensions besides the risk of death.

Comparison of Cox and Gray's survival models in severe sepsis.

BACKGROUND: Although survival is traditionally modeled using Cox proportional hazards modeling, this approach may be inappropriate in sepsis, in which the proportional hazards assumption does not hold. Newer, more flexible models, such as Gray's model, may be more appropriate. OBJECTIVES: To construct and compare Gray's model and two different Cox models in a large sepsis cohort. To determine whether hazards for death after sepsis were nonproportional. To explore how well the different survival modeling approaches describe these data. DESIGN: Analysis of combined data from the treatment and placebo arms of a large, negative, sepsis trial. SETTING: Intensive care units at 136 U.S. medical centers. SUBJECTS: A total of 1090 adults aged 18 yrs or older with signs and symptoms of severe sepsis and documented or probable Gram-negative infection. MEASUREMENTS: We considered 27 potential baseline risk factors and modeled survival over the 28 days after the onset of sepsis. We tested proportionality in single-variable Cox analysis using Schoenfeld residuals and log-log plots. We constructed a traditional multivariable Cox model, a multivariable Cox model with time-varying covariates, and a multivariable Gray's model. RESULTS: In single-variable analyses, 20 of the 27 potential factors were significantly associated with mortality, and 10 of 20 had nonproportional hazards. In multivariate analysis, all three models retained a very similar set of significant covariates (two models retained the identical set of nine variables, and the third differed only in that it retained the same nine plus a tenth variable). Four of the nine common covariates had nonproportional hazards. Of the three models, Gray's model best captured these changing hazard ratios over time. CONCLUSION: We confirm that many of the important predictors of mortality in severe sepsis are nonproportional and find that Gray's model seems best suited for modeling survival in this condition.

Variability of inspiratory conductance quantifies flow limitation.

In the present study, our aim was to investigate whether the variability of conductance over the course of inspiration reflects flow limitation. Pressure/flow conditions in the upper airway were modelled by a collapsible tube within a rigid chamber and a pump simulating respiration. Instantaneous conductance was estimated every 20 ms as flow/resistive pressure, and its variability during inspiration expressed as the 90th/50th percentile ratio. Accuracy of this ratio to quantify flow limitation was evaluated by observing whether it changed predictably with adjustments of model parameters. To illustrate the potential of this ratio to quantify flow limitation in a clinical setting, we recorded pneumotachographic airflow and oesophageal pressure in 11 patients with obstructive sleep apnoea during nasal continuous positive airway pressure (CPAP) ventilation, and observed changes in the 90th/50th percentile ratio of inspiratory lung conductance induced by mask pressure titration. Rising pressure surrounding the collapsible tube from subatmospheric to positive values induced progressive inspiratory collapse and increased 90th/50th percentile ratios of inspiratory conductance as predicted. Changes in flow limitation induced by other model modifications were also correctly tracked by the 90th/50th conductance percentile ratio. Increasing mask pressure during CPAP ventilation in sleep apnoea patients from subtherapeutic to therapeutic pressure levels was associated with the expected decrease in the 90th/50th percentile ratio of inspiratory lung conductance from a mean of 6.5+/-3.1 to 1.6+/-0.3 ( P <0.001). We conclude that variability of inspiratory conductance quantified by the 90th/50th percentile ratio may serve as a measure of flow limitation that is independent of the absolute value of conductance.