A human factors approach to observation chart design can trump health professionals’ prior chart experience

Aim

To determine whether experienced health professionals recognise patient deterioration more accurately and efficiently using (a) novel observation charts, designed from a human factors perspective, or (b) chart designs with which they have long-term experience.

Methods

Participants were 101 health professionals experienced in using either a multiple parameter track-and-trigger chart or a graphical chart with no track-and-trigger system. Participants were presented with realistic abnormal and normal patient observations recorded on six hospital observation charts of varying design quality, including the chart that participants were familiar with (or a very similar design). Across 48 trials, the participant was asked to specify if any of the vital sign observations were abnormal, or if all of the observations were normal. Participants’ overall error rates (i.e., proportion of incorrect responses) and response times, the main outcome measures, were calculated for each observation chart.

Results

Participants made significantly fewer errors and responded significantly faster when using a novel user-friendly chart compared with all the other designs, including the charts that they were experienced with in a clinical setting.

Conclusions

The findings suggest that, at least in the contexts examined, superior observation chart design appears to trump familiarity. Hence, hospitals motivated to improve the detection of patient deterioration should implement charts designed from a human factors perspective, rather than simply maintaining the status quo of reliance on clinical experience.     

CREWS: Improving specificity whilst maintaining sensitivity of the National Early Warning Score in patients with chronic hypoxaemia

Background

The National Early Warning Score (NEWS) is being introduced across the UK, but there are concerns about its specificity in patients with chronic hypoxaemia, such as some patients with COPD. This could lead to frequent clinically insignificant triggers and alarm fatigue.

Aims of study

To investigate whether patients with chronic hypoxaemia trigger excessively with NEWS, and to design a simple variant of NEWS for patients with chronic hypoxaemia: a Chronic Respiratory Early Warning Score (CREWS).

Methods

Data was collected from respiratory wards at two hospitals in North Wales. Components of NEWS and frequency of trigger thresholds being reached were recorded. CREWS was applied retrospectively to patients’ observations.

Results

196 admissions were analysed, including 78 for patients with chronic hypoxaemia. Patients with chronic hypoxaemia frequently exceeded trigger thresholds using NEWS during periods of stability/at discharge. Using CREWS, triggers during stability/at discharge were reduced from 32% of observations to 14% using a trigger threshold of a score greater than 6, and from 50% to 18% using a score greater than 5. All patients with chronic hypoxaemia who died within 30 days still reached CREWS trigger thresholds, and the area under receiver operated curves for NEWS and CREWS was comparable.

Conclusion

CREWS is a simple variant of NEWS for patients with chronic hypoxaemia that could reduce clinically insignificant triggers and alarm fatigue, whilst still identifying the sickest patients.     

Predicting clinical deterioration in the hospital: The impact of outcome selection

Background

Clinical deterioration of ward patients can result in intensive care unit (ICU) transfer, cardiac arrest (CA), and/or death. These different outcomes have been used to develop and test track and trigger systems, but the impact of outcome selection on the performance of prediction algorithms is unknown.

Methods

Patients hospitalized on the wards between November 2008 and August 2011 at an academic hospital were included in the study. Ward vital signs and demographic characteristics were compared across outcomes. The dataset was then split into derivation and validation cohorts. Logistic regression was used to derive four models (one per outcome and a combined outcome) for predicting each event within 24 h of a vital sign set. The models were compared in the validation cohort using the area under the receiver operating characteristic curve (AUC).

Results

A total of 59,643 patients were included in the study (including 109 ward CAs, 291 deaths, and 2638 ICU transfers). Most mean vital signs within 24 h of the events differed statistically, with those before death being the most deranged. Validation model AUCs were highest for predicting mortality (range 0.73–0.82), followed by CA (range 0.74–0.76), and lowest for predicting ICU transfer (range 0.68–0.71).

Conclusions

Despite differences in vital signs before CA, ICU transfer, and death, the different models performed similarly for detecting each outcome. Mortality was the easiest outcome to predict and ICU transfer the most difficult. Studies should be interpreted with these differences in mind.     

Risk prediction of hospital mortality for adult patients admitted to Australian and New Zealand intensive care units: Development and validation of the Australian and New Zealand Risk of Death model

Purpose

The purpose of this study is to develop and validate a new mortality prediction model (Australian and New Zealand Risk of Death [ANZROD]) for Australian and New Zealand intensive care units (ICUs) and compare its performance with the existing Acute Physiology and Chronic Health Evaluation (APACHE) III-j.

Materials and Methods

All ICU admissions from 2004 to 2009 were extracted from the Australian and New Zealand Intensive Care Society Adult Patient Database. Hospital mortality was modeled using logistic regression with training (two third) and validation (one third) data sets. Predictor variables included APACHE III score components, source of admission to ICU and hospital, lead time, elective surgery, treatment limitation, ventilation status, and APACHE III diagnoses. Model performance was assessed by standardized mortality ratio, Hosmer-Lemeshow C and H statistics, Brier score, Cox calibration regression, area under the receiver operating characteristic curve, and calibration curves.

Results

There were 456 605 patients available for model development and validation. Observed mortality was 11.3%. Performance measures (standardized mortality ratio, Hosmer-Lemeshow C and H statistics, and receiver operating characteristic curve) for the ANZROD and APACHE III-j model in the validation data set were 1.01, 104.9 and 111.4, and 0.902; 0.84, 1596.6 and 2087.3, and 0.885, respectively.

Conclusions

The ANZROD has better calibration; discrimination compared with the APACHE III-j. Further research is required to validate performance over time and in specific subgroups of ICU population.     

Safety and compliance with an emergency medical service direct psychiatric center transport protocol

Objectives

To evaluate compliance and safety of an emergency medical service (EMS) triage protocol that allows paramedics to transport patients directly to psychiatric emergency services.

Methods

A psychiatric patient diversion protocol was developed for our system. Protocol compliance was evaluated the following 3 ways: (1) psychiatric facility intake forms completed by mental health workers on patients transported by EMS directly to a psychiatric emergency service (PES) bypassing the ED, (2) hospital records for patients who were redirected from PES to the ED for medical evaluation, (3) retrospective analysis of ambulance charts. Study outcomes included protocol noncompliance rate, protocol failure rate, and any morbidity associated with either noncompliance or protocol failure. Data were analyzed using proportions and 95% confidence intervals (CI).

Results

A total of 174 patients were directly transported to PES bypassing ED medical clearance. The protocol effectively screened for medical issues in 96% of cases. Protocol noncompliance occurred in 51 cases for a frequency of 29% (CI, 22%-36%). One patient in the paramedic noncompliance group required hospital admission. There was protocol failure in 5 (2.9%; 95% CI, 0.9-6.6) of the patients who fit all protocol requirements for transport to PES but required secondary transport to the ED. All were subsequently transferred back to PES. Nine patients (5.2%; CI, 2.7%-9.5%) required secondary transfer to the ED. No patient had critical or life-threatening problems.

Conclusions

Emergency medical service providers showed a poor level of compliance with vital sign criteria, but the protocol provided a high level of safety.     

The association between vital signs and major hemorrhagic injury is significantly improved after controlling for sources of measurement variability

Purpose

Measurement error and transient variability affect vital signs. These issues are inconsistently considered in published reports and clinical practice. We investigated the association between major hemorrhagic injury and vital signs, successively applying analytic techniques that excluded unreliable measurements, reduced transient variation, and then controlled for ambiguity in individual vital signs through multivariate analysis.

Methods

Vital sign data from 671 adult prehospital trauma patients were analyzed retrospectively. Computer algorithms were used to identify and exclude unreliable data and to apply time averaging. An ensemble classifier was developed and tested by cross-validation. Primary outcome was hemorrhagic injury plus red cell transfusion. Areas under receiver operating characteristic curves (ROC AUCs) were compared by the test of DeLong et al.

Results

Of initial vital signs, systolic blood pressure (BP) had the highest ROC AUC of 0.71 (95% confidence interval, 0.64-0.78). The ROC AUCs improved after excluding unreliable data, significantly for heart rate and respiratory rate but not significantly for BP. Time averaging to reduce temporal variability further increased AUCs, significantly for BP and not significantly for heart rate and respiratory rate. The ensemble classifier yielded a final ROC AUC of 0.84 (95% confidence interval, 0.80-0.89) in cross-validation.

Conclusions

Techniques to reduce variability in vital sign data can lead to significantly improved diagnostic performance. Failure to consider such variability could significantly reduce clinical effectiveness or confound research investigations.     

Predictors of nurse manager stress: A dominance analysis of potential work environment stressors

Background

Nurse managers have important but stressful jobs. Clinical or bedside nurse predictors of stress have been studied more frequently, but less has been done on work environment predictors for those in this first-line leadership role. Understanding the relative importance of those work environment predictors could be used to help identify the most fruitful areas for intervention, potentially improving recruitment and retention for nurse managers.

Objective

Using Role Stress Theory and the Job Demands-Resources Theory, a model was tested examining the relative importance of five potential predictors of nurse manager stress (i.e., stressors). The work environment stressors included role ambiguity, role overload, role conflict, organizational constraints, and interpersonal conflict.

Design and settings

A quantitative, cross-sectional survey study was conducted with a convenience sample of 36 hospitals in the Southwestern United States.

Participants

All nurse managers working in these 36 hospitals were invited to participate. Of the 636 nurse managers invited, 480 responded, for a response rate of 75.5%.

Methods

Questionnaires were distributed during nursing leadership meetings and were returned in person (in sealed envelopes) or by mail.

Results

Because work environment stressors were correlated, dominance analysis was conducted to examine which stressors were the most important predictors of nurse manager stress. Role overload was the most important predictor of stress, with an average of 13% increase in variance explained. The second- and third-most important predictors were organizational constraints and role conflict, with an average of 7% and 6% increase in variance explained, respectively.

Conclusion

Because other research has shown deleterious effects of nurse manager stress, organizational leaders are encouraged to help nurse managers reduce their actual and/or perceived role overload and organizational constraints.     

Motivations and barriers to implementing electronic health records and ED information systems in Japan

Background

Although electronic health record systems (EHRs) and emergency department information systems (EDISs) enable safe, efficient, and high-quality care, these systems have not yet been studied well. Here, we assessed (1) the prevalence of EHRs and EDISs, (2) changes in efficiency in emergency medical practices after introducing EHR and EDIS, and (3) barriers to and expectations from the EHR-EDIS transition in EDs of medical facilities with EHRs in Japan.

Materials and methods

A survey regarding EHR (basic or comprehensive) and EDIS implementation was mailed to 466 hospitals. We examined the efficiency after EHR implementation and perceived barriers and expectations regarding the use of EDIS with existing EHRs. The survey was completed anonymously.

Results

Totally, 215 hospitals completed the survey (response rate, 46.1%), of which, 76.3% had basic EHRs, 4.2% had comprehensive EHRs, and 1.9% had EDISs. After introducing EHRs and EDISs, a reduction in the time required to access previous patient information and share patient information was noted, but no change was observed in the time required to produce medical records and the overall time for each medical care. For hospitals with EHRs, the most commonly cited barriers to EDIS implementation were inadequate funding for adoption and maintenance and potential adverse effects on workflow. The most desired function in the EHR-EDIS transition was establishing appropriate clinical guidelines for residents within their system.

Conclusion

To attract EDs to EDIS from EHR, systems focusing on decreasing the time required to produce medical records and establishing appropriate clinical guidelines for residents are required.     

Transbuccal fentanyl for rapid relief of orthopedic pain in the ED

Objectives

This study's objectives were to assess administration of a rapidly dissolving transbuccal fentanyl tablet to patients in emergency department (ED) with orthopedic extremity pain. The main end point was time required to achieve a 2-point drop on a 0 to 10 pain scale.

Methods

In this double-blind trial, subjects received either transbuccal fentanyl, 100 μg, and a swallowed placebo, or a swallowed oxycodone/acetaminophen, 5/325-mg pill, and a nonanalgesic transbuccal comparator. Pain assessment occurred every 5 minutes for an hour, and vital signs were monitored for 2 hours.

Results

Transbuccal fentanyl was associated with faster pain relief onset (median, 10 vs 35 minutes; P < .0001). Secondary end points (pain relief magnitude, rescue medication rate, subject preference for medication on future visit) favored transbuccal fentanyl. No vital sign abnormalities or significant side effects occurred in the ED or on 100% next-day follow-up.

Conclusions

Transbuccal fentanyl shows promise for continued investigation as a means to safely provide rapid and effective pain relief for ED patients.     

The occurrence of adverse events potentially attributable to nursing care in medical units: Cross sectional record review

Background

Ensuring the safety of hospitalized patients remains a major challenge for healthcare systems, and nursing services are at the center of hospital care. Yet our knowledge about safety of nursing care is quite limited. In fact, most earlier studies examined one, or at most two, indicators, thus presenting an incomplete picture of safety at an institutional or broader level. Furthermore, methodologies have differed from one study to another, making benchmarking difficult.

Objectives

The aim of this study was to describe the frequencies of six adverse events widely considered in the literature to be nursing-sensitive outcomes and to estimate the degree to which these events could be attributed to nursing care.

Method

Cross-sectional review of charts of 2699 patients hospitalized on 22 medical units in 11 hospitals in Quebec, Canada. The events included: pressure sores, falls, medication administration errors, pneumonias, urinary infections, and inappropriate use of restraints. Experienced nurse reviewers abstracted patients’ charts based on a grid developed for the study.

Results

Patient-level risk for at least one of these six adverse events was 15.3%, ranging from 9% to 28% across units. Of the 412 patients who experienced an event, 30% experienced two or more, for a total of 568 events. The risk of experiencing an adverse event with consequences was 6.2%, with a unit-level range from 3.2% to 13.5%. Abstractors concluded that 76.8% of the events were attributable to nursing care.

Conclusion

While the measurement approach adopted here has limitations stemming from reliance on review of documentation, it provided a practical means of assessing several nursing-sensitive adverse events simultaneously. Given that patient safety issues are so complex, tracking their prevalence and impact is important, as is finding means of evaluating progress in reducing them.     

Age- and sex-specific normal values for shock index in National Health and Nutrition Examination Survey 1999-2008 for ages 8 years and older

Purpose

Shock index (SI), the ratio of heart rate to systolic blood pressure, has found to outperform conventional vital signs as a predictor of shock. Although age-specific vital sign norms are recommended in screening for shock, there are no reported age- or sex-specific norms for SI. Our primary goal was to report age- and sex-specific SI normal values for a nationally representative population 10 years and older by 5-year age groups. A secondary goal was to report SI normal values for children ages 8 to 19 years by 1-year age groups.

Basic procedures

Weighted data from the National Health and Nutrition Examination Survey 1999-2008 data sets were used to generate age- and sex-specific percentile curves of SI for subjects 8 years and older.

Main findings

The primary analysis included 33 906 subjects (101 837 weighted) 10 years and older. The secondary analysis included 13 393 subjects (37 983 weighted) 8 to 19 years old. Normalized SI values for each percentile decreased with increasing age and were higher for females across all ages. The most commonly cited SI threshold of 0.9 exceeded the 97th percentile for males younger than 25 years and for females younger than 40 years.

Conclusions

This first report of age- and sex-specific normal values for SI indicates that SI norms vary by age and sex. Just as age-specific vital sign norms are recommended in screening for shock, our findings suggest that age- and sex-specific SI norms may be more effective in screening for shock than a single-value threshold.     

Portable miniaturized extracorporeal membrane oxygenation systems for H1N1-related severe acute respiratory distress syndrome: A case series

Background

Technological advances improved the practice of “modern” extracorporeal membrane oxygenation (ECMO). In the present report, we describe the experience of a referral ECMO center using portable miniaturized ECMO systems for H1N1-related severe acute respiratory distress syndrome (ARDS).

Methods

An observational study of all patients with H1N1-associated ARDS treated with ECMO in Hospital S. João (Porto, Portugal) between November 2009 and April 2011 was performed. Extracorporeal membrane oxygenation support was established using either ELS or Cardiohelp systems (Maquet-Cardiopulmonary-AG, Hirrlingen, Germany).

Results

Ten adult patients with severe ARDS secondary to H1N1 infection (Pao₂/fraction of inspired oxygen, 69 mm Hg [56-84]; Murray score, 3.5 [3.5-3.8]) were included, and 60% survived to hospital discharge. Five patients were uneventfully transferred on ECMO from referring hospitals to our center by ambulance. Six patients were treated during the first postpandemic influenza season. All patients were treated with oseltamivir, and 1 received in addition zanamivir. Four patients received corticosteroids. Nosocomial infection was the most common complication (40%). Of the 4 deaths, 2 were caused by hemorrhagic shock; 1, by irreversible multiple organ failure; and 1, by refractory septic shock.

Conclusion

In our experience, ECMO support was a valuable therapeutic option for H1N1-related severe ARDS. The use of portable miniaturized systems allowed urgent rescue of patients from referring hospitals and safe interhospital and intrahospital transport during ECMO support.     

Expansion of a medical emergency team system to a mental health facility

Background

Clinical emergency response systems such as medical emergency teams (MET) are used in many hospitals worldwide, but the effect that these systems have in mental health facilities is unknown. This study examined the rate and nature of MET calls to a mental health facility that had relocated to the campus of a tertiary referral hospital.

Methods

This study was a prospective, observational study of MET calls to a newly constructed 170 bed mental health facility. Data were collected on the number and nature of MET calls to the facility.

Results

Over 24 months, there were 66 MET calls to the mental health facility, and 1217 MET calls at the main hospital. The mean MET call rate was 14.2 calls per 1000 admissions (95% confidence interval (CI) 10.8–17.7) at the mental health facility, and 14.7 calls per 1000 admissions (95% CI 13.9–15.5) at the main hospital. Neurological and cardiovascular problems were present in 61% and 41% of MET calls.

Conclusion

The rate of MET calls to a new mental health facility can be similar to that of a tertiary hospital. Staff attending MET calls need to be prepared to manage predominantly neurological and cardiovascular problems.     

Olanzapine in ED patients: differential effects on oxygenation in patients with alcohol intoxication

Introduction

Agitation has significant consequences for patients and staff. When verbal techniques fail, expert guidelines recommend the use of second-generation antipsychotics (SGAs). Perhaps out of familiarity with haloperidol and benzodiazepines, emergency department (ED) clinicians often pair SGAs with benzodiazepines as well. Use of SGAs such as olanzapine in alcohol-intoxicated (ETOH+) patients or with benzodiazepines is not well studied and may be associated with vital sign abnormalities.

Methods

This is a structured chart review of all patient visits who received either oral or intramuscular (IM) olanzapine in an academic ED from 2004 to 2010 and who had systolic blood pressure, heart rate, and oxygen saturation documented before medication administration and within 4 hours afterwards.

Results

Four hundred eighty-two patient visits received olanzapine; 275 patient visits (225 oral, 50 IM) had vital signs documented. Neither route of administration, concurrent benzodiazepines, nor ingestion of ETOH were associated with significant decreases in systolic BP or heart rate (P = ns for all comparisons). Decreases in oxygen saturations, however, were significantly larger in ETOH+ patients who received IM olanzapine or IM olanzapine + benzodiazepines. Route of administration, concurrent benzodiazepines, nor ingestion of ETOH was associated with significant decreases in systolic blood pressure or heart rate (p = ns for all comparisons). Decreases in oxygen saturations, however, were significantly larger in ETOH+ patients who received IM olanzapine or IM olanzapine + benzodiazepines.

Conclusions

Oral olanzapine was not associated with significant vital sign changes in ED patients. Intramuscular olanzapine also was not associated with vital sign changes in ETOH− patients. In ETOH+ patients, IM olanzapine was associated with significant oxygen desaturations. In ETOH+ ED patients, oral olanzapine (with or without benzodiazepines) or haloperidol may be safer choices. ETOH+ patients may have differential effects with the use of IM SGAs such as olanzapine and should be studied separately in drug trials.     

Medical emergency team response for the non-hospitalized patient

Objectives

Rapid response systems (RRS) evolved to care for deteriorating hospitalized patients outside of the ICU. However, emergent critical care needs occur suddenly and unexpectedly throughout the hospital campus, including areas with non-hospitalized persons. The efficacy of RRS in this population has not yet been described or tested. We hypothesize that non-hospitalized patients accrue minimal benefit from ICU physician participation in the RRS.

Design

A retrospective review of all RRS events in non-hospitalized patients for a 28 month period was performed in a large, urban university medical center. Location, patient type and age, activation trigger, interventions performed, duration of event and disposition were recorded. Admission diagnosis and length of stay were also recorded for patients admitted to the hospital.

Setting

Academic medical center.

Patients

Non-hospitalized persons requiring evaluation by the medical emergency team.

Interventions

None.

Measurements and main results

There were a total of 1778 RRS activations during the study period. 232 (13%) of activations were for non-hospitalized patients. The patient cohort consisted of outpatients, visitors, and staff. Triggers for RRS activation were neurologic change (42%), cardiac (27%), respiratory (16%), and staff concerns (16%). The mean duration of the response was 38 min. The most common interventions performed included administration of oxygen (46%), intravenous fluids (13%) and dextrose (6%). 82% of patients were taken to the emergency department and 32% of the ED cohort were admitted to the hospital.

Conclusions

Perceived emergencies in non-hospitalized patients occur commonly but require minimal emergent intervention. Restriction of critical care physician involvement to inpatient deteriorations should be considered when designing a RRS. Future studies are needed to evaluate the utility of non-physician provider led rapid response teams with protocol-driven interventions for similar populations.     

Documentation of code status and discussion of goals of care in gravely ill hospitalized patients

Background

Timely discussions about goals of care in critically ill patients have been shown to be important.

Methods

We conducted a retrospective chart review over 2 years (2003-2004) of patients admitted to our medical service who were classified as “expected to die.” Charts were evaluated for do-not-resuscitate (DNR) documentation and discussions of goals of care. Detailed chart reviews for demographic information, cause of death, site of death, length of stay, and duration of resuscitation attempt were performed.

Results

Of 497 charts identified, 434 (87.3%) had a DNR on file at the time of death. After exclusion of patients who died in less than 24 hours, 18 no-DNR charts remained. Seven noted a decision to continue aggressive care and 11 had no code status discussion documented. Younger patients and patients with cardiovascular disease were less likely to have a DNR. Resuscitation times were longer in the no-discussion group. All patients who died without a DNR died in the intensive care unit. Seventy-six percent of discussions were done by medicine housestaff.

Conclusions

Although the overall rate of DNR documentation was high, several trends emerged. Medicine housestaff in the intensive care unit would be a logical group to target for an educational intervention to address these discrepancies.     

Intramuscular Ziprasidone: Influence of Alcohol and Benzodiazepines on Vital Signs in the Emergency Setting

Background

Ziprasidone is a second-generation antipsychotic (SGA) approved for agitation. Few previous studies have examined ziprasidone in the emergency department (ED). For instance, it is unknown how often emergency physicians prescribe ziprasidone, whether it is typically prescribed in combination with a benzodiazepine, or whether use of intramuscular (i.m.) ziprasidone and benzodiazepines affects vital signs compared to i.m. ziprasidone alone. Objective: Our aims were to determine the demographics of patients receiving ziprasidone in an urban-suburban ED; the relative frequency with which ziprasidone is prescribed; and the effects on vital signs, repeat medication dosage, and lengths of stay.

Methods

This is a multicentered structured chart review from 2003 to 2010 of ziprasidone use at two hospitals. If documented, vital signs were compared in patients who received concurrent benzodiazepines and in those who did not, and in patients who ingested alcohol and in those who did not.

Results

Patients on 95 visits received ziprasidone during the study period, with one third of these receiving accompanying benzodiazepines. Forty-nine unique patients who were treated with i.m. ziprasidone had documented vital signs. In these patients, alcohol intoxication was associated with decreased oxygen saturations irrespective of benzodiazepines. Concurrent benzodiazepines had no other deleterious effect on vital signs but resulted in longer ED stays.

Conclusions

This study suggests that many ED physicians, who commonly prescribe a benzodiazepine with a first-generation antipsychotic like haloperidol, have transferred this practice to SGAs like ziprasidone. In this sample, this pairing did not adversely affect vital signs but was associated with marginally longer ED stays. Caution should be exercised when treating alcohol-intoxicated patients with ziprasidone, as this can decrease oxygen saturations.     

Etiology of troponin elevation in critically ill patients

Purpose

The aim of this study was to assess the etiology of cardiac troponin elevation among patients admitted to the intensive care unit (ICU) and to examine whether etiology affects mortality and length of stay.

Methods

All patients admitted over 2 months underwent screening with troponin measurements and were included if 1 or more measurements were elevated. Two adjudicators retrospectively reviewed patient charts to determine the likely cause of troponin elevation.

Results

Of 103 patient admissions, 52 (50.5%) had 1 or more elevated troponin measurements, and 49 (94.2%) had medical charts available for review. Troponin elevation was adjudicated as myocardial infarction (MI) in 53.1% of patients, sepsis in 18.4%, renal failure in 12.2%, and other causes in 16.3%. Overall ICU mortality was 16.0%; 2.0% for patients with no troponin elevation, 23.1% in patients with MI, and 39.1% in patients with troponin elevation not due to MI. Having an elevated troponin level not due to MI was significantly associated with increased hospital mortality compared with having no troponin elevation.

Conclusions

The most common cause of troponin elevation among critically ill patients was MI. Patients with elevated troponin had worse outcomes compared with patients without troponin elevation, and troponin elevation not due to MI was predictive of increased hospital mortality.     

Prehospital use of continuous positive airway pressure for acute severe congestive heart failure

Background

The utility of continuous positive airway pressure (CPAP) in the in-hospital treatment of congestive heart failure (CHF) is well established. Its exact role is less clear in the prehospital arena.

Objective

To describe the prehospital use of CPAP for patients presenting with acute severe heart failure in a large Emergency Medical Services system in New Jersey.

Methods

Retrospective review of prehospital charts from January 1, 2005 to December 31, 2006 of patients treated for acute CHF. Inclusion criteria for eligibility for CPAP mask use were: respiratory rate > 25 breaths/min, labored and shallow breathing, bilateral rales, history of CHF, intact mental status, and prehospital clinical diagnosis of CHF. Data collected included demographics, vital signs, oxygen saturation (SaO₂), need for endotracheal intubation (ETI), and complications.

Results

There were 1306 charts reviewed; 387 patients met inclusion criteria. Of the 387, 149 patients had placement of CPAP (38.5%). The prehospital treatment times were (CPAP = 30 min; non-CPAP = 31 min; p < 0.01). The increase in SaO₂ for the CPAP group (9%) vs. the non-CPAP group (5%) was statistically significant (p < 0.01). Systolic blood pressure (BP) reduction (CPAP [27.1 mm Hg], non-CPAP [19.9 mm Hg], p < 0.01), diastolic BP reduction (CPAP [14.1 mm Hg], non-CPAP [7.4 mm Hg], p < 0.01), heart rate reduction (CPAP [17.2 beats/min], non-CPAP [9.6 beats/min], p < 0.01), respiratory rate reduction (CPAP [5.63], non-CPAP [4.09], p < 0.01), and ETI reduction (CPAP [2.6%], non-CPAP [5.46%], p < 0.01), all were statistically significant. Adjunctive CHF treatments were similar between the groups.

Conclusion

The use of CPAP for eligible patients with acute severe CHF seems to be feasible and beneficial. Large-scale randomized prospective prehospital studies are needed to validate these results.