Objective We sought to identify the characteristics of patients who experience medical emergency team calls in the radiology department (MET-RD) and the relationship between these characteristics and patient outcomes.
Design/participants Retrospective review of 111 inpatient MET-RD calls (May 2008–April 2010).
Setting Academic medical centre with a well established MET system.
Measurements The characteristics of patients before, during and after transport to radiology were extracted from medical records and administrative databases. These characteristics were compared between patients with good and poor outcomes.
Main results The majority of patients who experience MET-RD calls had a Charlson Comorbidity Index ≥4 and were from non-intensive care units (60%). Almost half (43%) of MET-RD calls occurred during patients' first day in hospital. Patients commonly arrived with nasal cannula oxygen (38%), recent tachypnoea (28%) and tachycardia (34%). A minority (16%) fulfilled MET call criteria in the 12 h before the MET-RD. MET-RD etiologies were cardiac (41%), respiratory (29%) or neurological (25%), and occurred most frequently during CT (44%) and MRI (22%) testing. Post MET-RD, the majority of patients (70%) required a higher level of care. Death before discharge (25%) was associated with need for cardiovascular support prior to RD transport (p=0.02), need for RD monitoring (p=0.02) and need for heightened RD surveillance (p=0.04).
Conclusions The majority of patients who experienced MET-RD calls came from non-intensive care units, with comorbidities and vital sign alterations prior to arrival at the RD. Risk appeared to be increased for those requiring CT and MRI. These findings suggest that prior identification of a subset of patients at risk of instability in the RD may be possible.
- Patient safety
- medical emergency team
- patient outcomes
- intra-hospital transport
- patient safety
- hospital medicine
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
- Patient safety
- medical emergency team
- patient outcomes
- intra-hospital transport
- patient safety
- hospital medicine
Medical emergency teams (METs) have emerged as a means to provide critical care expertise when additional resources are needed for the management of compromised patients outside the intensive care unit (ICU).1–4 METs represent the efferent or response arm of a rapid response systems (RRS). Despite mixed findings on the efficacy of METs,1 5 6 the Institute of Healthcare Improvement's 100 000 lives campaign recommends that institutions implement a RRS as a means to reduce potentially preventable deaths.7 Additionally, the Joint Commission includes RRS implementation in their 2008 Patient Safety Goals.8
Prior studies describe MET activation and outcomes in various hospital settings.1–4 6 9 However, few studies have examined MET utilisation in the radiology department (RD). Inpatients are frequently transported for diagnostic tests and procedures, and some of these patients become unstable and require a MET call (MET-RD).10–13 Early recognition of patient instability and complications in the RD may be hindered by the wide range of patient conditions and acuity levels seen in this department. At least one study found that service/unit mismatch, that is, care not provided on a designated unit or by staff not typically assigned to the unit, was predictive of more MET calls.14 This finding highlights the potential vulnerability of patients cared for by nurses in a setting, such as the RD.
There is literature to demonstrate that patients may exhibit signs of physiological compromise before healthcare professionals have recognised a problem and initiated patient rescue or called the MET.14–18 There is also literature demonstrating that patients may become unstable during transport or require transport to determine the cause of instability.11 19 However, little is known about instability incidence and care of patients after they arrive in the clinical care area of the RD. Better characterisation of patients prior to a MET-RD call may aid in more appropriate determination of risk factors and preemptive determination of surveillance needs. Therefore, the purpose of this exploratory descriptive study was to identify characteristics of inpatients who experience MET-RD activation, and the relationship between these characteristics and patient outcomes.
Materials and methods
After obtaining institutional review board approval, we performed a retrospective review of all MET-RD activations for patients aged 18 years and older between 1 May 2008 and 30 April 2010 at a tertiary care academic medical centre with a well established MET system. The RD at this hospital performs diagnostic tests and procedures, including CT, MRI, interventional radiology, nuclear medicine, ultrasound, fluoroscopy and general reontenographical studies for approximately 160 000 inpatients annually. Eligible patients were identified through the facility's Medical Emergency Response Improvement Team (MERIT) database, which maintains data on all MET calls occurring in the study hospital. All visits for patients who experienced a MET-RD call during the study period were reviewed. Patients who experienced a MET activation while in transit to or from the RD were excluded from the study, as were MET-RD calls for outpatients (eg, ambulatory testing). The final sample consisted of 111 MET calls.
The RD staff members use the same MET trigger criteria (box 1) to respond to patient instability as staff on the general hospital units. The standard of care is to call the MET for all patients (non-ICU and ICU) who meet MET criteria outside of the ICU. Any nurse, physician, patient or visitor is able to call the MET by dialling a predetermined, well publicised number. The MET team consists of a critical care medicine physician, two critical care nurses and a respiratory therapist.
University of Pittsburgh Medical Center medical emergency team call criteria during study period.1
Rate <8 or >36
New onset of difficult breathing
New pulse oximeter reading <85% for >5 min without known chronic hypoxia
Heart Rate <40 or >140 with new symptoms or any rate >160
Blood pressure <80 or >200 systolic or 110 diastolic with symptoms (neurological changes, angina, dyspnoea)
Acute neurological change
Acute loss of consciousness
New onset lethargy
Seizure (outside of seizure monitor unit)
Sudden loss of movement or weakness in the face, arm or leg
>1 emergency page required to assemble team needed to respond to a crisis
Patient complaint of chest pain (unresponsive to nitroglycerine or doctor unavailable)
Colour change (patient or extremity): pale, dusky, grey or blue
Unexplained agitation of >10 min
Narcan use without immediate response
New oxygen requirement to keep oxygen saturation >85%
Bleeding into airway
Large acute blood loss
Crash cart must be used for rapid delivery of meds
Data collection and variables of interest
We recorded the number of MET-RD calls and the time of day and day of week when they occurred. Time of day was categorised by 4 h increments for analysis. The characteristics of patients before RD transport (demographics, admitting diagnosis by International Classification of Disease ninth revision (ICD-9) code, Charlson Comorbidity Index (CCI) individual and total scores,20 transfer from ICU or non-ICU unit of origin, care needs and vital signs in the 12 h before RD arrival) and during RD care (RD modality, RD care needs, type of RD caregivers) were extracted from the electronic medical record or the administrative database.20 Admitting ICD-9 codes were categorised into eight related categories representing primary diagnosis.
The care needs in the 12 h prior to RD transport were identified from the patient record. Respiratory support was defined as the need for any supplemental oxygen delivery system or type of respiratory support (nasal cannulae, face mask, mechanical ventilation). Cardiovascular support was defined as documentation of inotropic medications and/or fluid resuscitation and need for blood products. Cardiac arrhythmias were identified using cardiac assessments documented by nurses. Sedation prior to arrival at the RD was recorded and included propofol, diazepam, morphine, fentanyl, alprazolam, hydromorphone, midazolam, succinylcholine or vecuronium. Vital signs were categorised according to MET call criteria (box 1). In addition, tachycardia was defined as a heart rate >100 beats/min, and tachypnoea on cut-off points of rate ≥24 breaths/min and ≥30 breaths/min.
The RD modalities at the time of the MET-RD call and the aetiology of the call were obtained from the MET call records. MET aetiology was classified as a cardiac, respiratory, neurological or other cause. Sedation in the RD consisted of administration of lorazepam, hydromorphone, midazolam, fentanyl, propofol, succinylcholine or vecuronium. Monitoring in the RD consisted of any combination of cardiac, blood pressure and pulse oxymetry. The level of RD surveillance (the type of healthcare professional monitoring the patient) was described according to four caregiver categories: RD technician (responsible for conducting the test ordered for the patient); staff registered nurse (RN, general ward, resource pool or ICU staff) who accompanied the patient to the RD; RD RN (RD-unit assigned); or advanced patient care technician (APCT, unlicensed care provider trained in reading cardiac monitors who accompanies monitored patients during transport within the hospital).
Two separate patient outcomes were examined. The first was whether or not patients required a more intense level of care post MET-RD. ‘Higher level of care’ was considered to exist when patients experienced any of the following: increased respiratory support (nasal cannulae, high flow oxygen, mechanical ventilation); increased cardiac support (fluid resuscitation, blood products, pharmacological blood pressure support or pharmacological cardiac rhythm support); emergent diagnostic or surgical procedure (angiography, cardiac cath lab, central venous catheterisation, EEG, electrocardiogram, CT, MRI, operating room); transfer to a higher acuity patient care unit. The second patient outcome examined was death between the MET-RD call and discharge. Study data were extracted by the study's principal investigator (LKO) and a graduate research assistant. To assess data collector quality control, 10% of records were reviewed separately by both data collectors and values for study variables compared, with no discrepancies noted between data collectors.
Statistical analysis was performed using SAS 9.2. Continuous variables were reported as mean±SD, and comparisons were made using univariate binary logistic regression. Categorical variables were reported as frequencies with percentages, and comparisons were made using χ2 tests and Fisher's exact test. Statistical significance was set at a p value of <0.05.
The distribution of the 111 MET-RD calls across day of week and time of day are illustrated in figure 1. MET-RD calls occurred more frequently in the middle of the week (Wednesday) and between 08:00 and 12:00 (30%).
Over 70% of patients required a higher level of care post MET-RD call (figure 2). Half required increased respiratory support and, of those not on mechanical ventilation prior to transport, 26% required mechanical ventilation. Of the 67 patients not admitted to an ICU prior to the MET-RD call, 38 were transferred to a higher acuity unit, 28 were newly admitted to an ICU, and 10 were sent to monitored non-ICU beds. Of the 26 patients who required an emergent procedure post MET-RD call, seven were sent to the operating room. In addition, 11 patients underwent further RD procedures (CT, angiography or MRI) post MET-RD call prior to leaving the RD. One in four patients (25%) who experienced a MET-RD call died prior to hospital discharge.
The characteristics of the patients prior to their RD transport were middle aged, Caucasian (81%) and with an equal distribution of women and men (table 1). Patients were evenly distributed across admitting diagnosis categories, with a slightly higher proportion with a neurological diagnosis (24%). The average CCI was approximately 4, with the most common comorbidities being renal (61%), cerebral vascular disease (28%), diabetes (22%), myocardial infarct (21%) and cardiopulmonary disease (20%). The majority of patients (60%) originated from non-ICUs. Almost half (43%) experienced the MET-RD call on their first day of hospitalisation. Many (65%) were on respiratory support, most commonly by nasal cannula oxygen (38%). In addition, prior to transport to the RD, 25% received cardiovascular support and 23% received sedation. Cardiac arrhythmias were documented in 27%. In the 12 h prior to the MET-RD call, 16% of patients had at some point met or exceeded MET call vital sign thresholds. While only 5% and 2% exceeded MET criteria for respirations and heart rate, respectively, 28% experienced tachypnoea (breaths/min ≥24) and 34% experienced tachycardia at levels below MET call thresholds (ie, between 100 and 160 beats/min).
When evaluating the association between patient characteristics prior to RD transport and the outcome of need for a higher level of care post MET (table 1), no significant differences were identified between patients requiring higher levels of care and those who did not. However, several trends were noted. Not surprisingly, non-ICU patients more often required a higher level of care (63%, p=0.43) as did patients who arrived on nasal cannula oxygen (28%, p=0.15). Of the patients with respirations ≥30 breaths/min, 92% required a higher level of care compared with 77% of patients with respirations ≥24 breaths/min.
In terms of patient characteristics prior to RD transport and associated inpatient mortality post MET-RD call (table 1), we observed a non-statistically significant trend towards increased mortality for men compared to women (68% vs 32%, p=0.09). A significantly greater proportion of the patients who died originated from an ICU (57%) than a ward (43%, p=0.03), and had been receiving cardiovascular support prior to arrival at the RD (39% vs 17%; p=0.02). However, a prior cardiac arrhythmia did not confer any significant mortality risk.
The majority of MET-RD calls were for patients receiving either CT (44%) or MRI (22%) scans (table 2). MET-RD aetiology was most commonly cardiac (41%) with the MET triggers of hypotension (17%), cardiac arrhythmias (16%) and chest pain (7%). Respiratory MET-RD aetiology (29%) triggers were hypoxia (23%) and airway protection (7%). Neurological MET-RD aetiology (25%) triggers were seizures (14%), altered mental status (9%) and stroke (1%). Other MET-RD triggers were falls, dislodged central venous catheters (2%) and the need for additional ICU personnel (2%).
There also appeared to be differences in surveillance. The majority of patients who experienced MET-RD calls (57%) received at least one form of continuous vital sign monitoring while in the RD; less than half were under the surveillance of a staff nurse (general ward, resource pool, ICU). The patients who were continuously monitored in the RD and those who were monitored by staff nurse surveillance in the RD were more likely to die during hospitalisation, suggesting prior identification of instability. Of the 18 patients (12 ICU and 6 non-ICU) with vital signs that met the MET criteria in the 12 h prior to the MET-RD call (table 1), 15 were monitored by an RN while in the RD, two patients were monitored by an APCT and one patient was not monitored.
Comparisons of the MET-RD sample by unit of origin are illustrated in table 3. The post MET-RD outcomes showed that patients from an ICU were more likely to die prior to discharge. There was no statistical difference in the proportion of patients from a non-ICU requiring a higher level of care post MET-RD call. The ICU patients were older, requiring greater respiratory and cardiac support with documented arrhythmias. Additionally, the ICU patients were more likely to have alterations in their vital signs and require sedation in the RD. There were no statistical differences in the MET-RD aetiologies and RD modalities.
Our study found variations in the time, unit of origin, type of testing and vital sign alterations among the 111 patients who experienced a MET-RD call over a two years period at a tertiary care academic hospital. To our knowledge, this is the largest study to date of patient instability in the RD. MET-RD calls were concentrated in the daylight hours and during weekdays. The majority of patients who experienced a MET-RD call arrived from non-ICUs. Patients having CT scans had the highest percentage of MET-RD calls. Patients often had documented alterations in vital signs, notably tachycardia and tachypnoea, in the 12 h prior to the MET-RD call. Over half of the patients who required a MET-RD call received continuous monitoring of vital signs while in the RD. A large majority (70%) required a higher level of care post MET-RD call and 25% died during hospitalisation post MET-RD call.
The post-MET-RD in-hospital death rate of 25% is somewhat higher than the estimates of 11–19% reported for inpatient mortality post MET calls to all patient care areas in other centres.6 10 This reflects the increased acuity of patients from ICUs cared for in the RD. The non-ICU post-MET-RD mortality rate (18%) was similar to the mortality rate reported in prior studies; however the ICU mortality rate (36%) was double the non-ICU rate. Interestingly, of the 70% of patients requiring a higher level of care, 34% required a higher acuity care unit and 23% required post-MET-RD procedures. The patients requiring a higher acuity care unit were all patients from non-ICUs; however, the patients requiring post-MET-RD procedures were equally divided between patients from ICUs and non-ICUs. Chan et al (2008), examining post-MET outcomes for general ward patients, reported only 46% required a higher level of care, but of these, 45% required a higher acuity care unit and only 0.6% required a procedure post MET. Patients from ICUs arrive in the RD already critically ill and potentially having been sent to the RD for investigation of deteriorating conditions. Despite the increased surveillance accompanying patients from ICUs to the RD, activation of the MET is needed in response to patient deterioration.21
CCI total scores ≥5 have been associated with high mortality and comorbid-related complications in hospitalised patients.22 23 CCI total scores (48% with scores ≥4) may also reflect on the 25% mortality seen in this sample. Compared with the inpatient population, CCI scores reported in other instability studies (8.9–20% of patients had a CCI ≥4),18 23 24 the MET-RD sample had an increased comorbiditiy burden. Given that there was no significant difference between the CCI scores of patients from ICUs and non-ICUs, the inclusion of patients from ICUs does not explain the increased comorbidity burden of the RD sample.
The diurnal variations in MET-RD calls noted in this study followed a pattern previously reported in the hospital-wide MET literature, with the highest number of calls from 08:00 to 12:00 h (representing the beginning of the day shift) and the majority of the calls being made on weekdays.14 18 25 The RD procedural units at our study site serve inpatients and outpatients, with outpatient procedures more heavily scheduled on weekdays. Care demands for outpatients on weekdays may have influenced this pattern; however, it may also reflect differences in characteristics of the inpatients receiving procedures during the week.
The most common radiological modality for a MET-RD call was CT (44%). In over half of patients who experienced a MET-RD call and died during hospitalisation (54%), the call was related to a CT procedure. This is consistent with findings from a prior study in this institution and with intra-hospital transport literature.21 26 CT has been identified as one of the most common types of procedures performed for inpatients and as a procedure involving patients at greatest risk for adverse events. From a review of eight studies examining 650 patients from ICUs who were transported for testing, 50% involved a CT scanner.19 There was no significant difference in RD modalities for patients from ICUs and non-ICUs in the MET-RD sample. Therefore, the documented high incidence of CT scans for patients from ICUs does not explain the high mortality for patients who experienced a MET-RD call during CT.
In a prospective study of 125 intra-hospital transports of patients from ICUs, one-third experienced an instability event that was potentially life threatening, and the majority of these events (75%) occurred in the RD or the operating room.11 Equipment failure has been cited as a cause of adverse events during intra-hospital transport of patients from ICUs to the RD; alterations in blood pressure, heart rate and oxygenation are also cited.27 Patients often show signs of physiological compromise hours before clinicians recognise the need for rescue interventions.15–18 Therefore, our examination of the recorded vital signs for all patients who experience a MET-RD in the 12 h prior to the call showed that 16% reached the MET call threshold of concern in one or more of the vital sign parameters (although a MET was not called). Perhaps more importantly, there were signs of physiological compromise prior to sending the patient to the RD, as evidenced by 34% of the patients with recorded heart rates >100 beats/min and 28% of the patients with recorded respiratory rates ≥24 breaths/min (11% ≥30 breaths/min). While not meeting the MET call threshold, these changes likely signalled physiological compromise. In prior studies, a respiratory rate ≥24 breaths/min has been associated with critical illness and ≥30 breaths/min has been predictive of higher acuity care needs and hospital mortality.28 29 Our study did not evaluate whether care was modified to manage these vital sign changes. Therefore, it is unknown whether preventive actions might have averted the MET-RD call. Patients from ICUs were statistically more likely to have alterations in vital signs, however this finding reinforces the need to closely monitor vital signs and the importance of this parameter as a signal of compromise.
Interestingly, the majority of patients who experienced a MET-RD call were on continuous electronic vital sign monitoring while in the RD. Staff RNs (general ward, resource pool or ICU) accompanied 43% of the patients to the RD, of whom only 8% were from non-ICUs. This suggests that the acuity of patients from non-ICUs may be underestimated prior to transfer to the RD, potentially resulting in decreased nursing surveillance in the RD. Among the patients who were not accompanied by a staff nurse to the RD, the technician responsible for conducting the radiological scan was the only source of patient surveillance for 40% of patients who experienced a MET-RD call. The RD RN (a member of the regular RD staff) provided surveillance for 14% of patients prior to the MET-RD call. In the study site, a 1:1 RD RN to patient ratio for direct nursing surveillance occurs only for select high-risk procedures (ie, conscious sedation, lung biopsy) or when specific medications are required by the institution's protocol for the procedure (ie, coronary artery CT angiography, cholescintigraphy). Although an RD RN can be requested to respond to increased surveillance needs, including instability, the technician or radiologist must specifically request this assistance. Our study design did not examine if the instability was of rapid onset or the recognition of instability by RD technicians. Direct nursing surveillance of all patients in the RD is neither practical nor necessary. However, the challenge is to identify compromise in patients prior to the need for a MET call to better allocate resources to provide for preemptive nursing surveillance. Our findings suggest that close monitoring of changes in heart and respiratory rate may provide an important initial warning.
In this study, the aetiology for the MET call was most commonly of a cardiac nature (41%), which may in part be due to the comorbidity prevalence of myocardial infarct (21%) and cardiopulmonary disease (20%) in the sample. Only a third of the MET call aetiology was due to a respiratory cause, while this is the most common reason for a MET call in patients from general wards.4 30 One explanation may be that, unlike the general ward MET studies, our sample includes patients from ICUs and 15% of our sample arrived in the RD already intubated and accompanied by a respiratory therapist as per hospital policy. However, there was no significant difference between the MET-RD aetiology of patients from ICUs and non-ICUs. Additionally, we did not find the type of MET-RD aetiology to be significantly associated with need for post-MET higher level of care or mortality.
The data are from a single tertiary academic medical centre with a well established MET system, which may limit the generalisability of these findings to the inpatient populations seen by other RDs. Also, despite the 2-year time frame, the study examined a relatively small sample of events. Data were collected retrospectively from various existing data sources. The data available for review did not provide information regarding specific RD tests, their emergent nature, exam outcomes or physician rationale for sending patients to the RD, all of which would have strengthened the study. Additionally, we did not include patients who experienced a MET in transit to the RD because the event was outside the influence of the RD. In addition, the volume of in-transit calls was very low. We recognise that clinical data sources, such as the medical record, are not designed for research purposes, and missing or inconsistently recorded data may introduce threats to validity. Finally, we did not describe our MET-RD daily prevalence relative to the number of procedures performed per time of day or day of the week. Thus, the patterns of MET-RD calls in the morning hours and mid week may be proportional to work flow. Our administrative dataset did not allow us to determine the time-specific workload. The findings, nevertheless, suggest the need for further prospective studies in radiological settings.
The characteristics and outcomes of inpatients who experience a MET call while in the RD may differ from those of inpatients who experience a MET call on a general ward, with regard to MET aetiology and comorbidity burden. Patients who experience a MET-RD call are from ICUs and non-ICUs, and therefore they are potentially more clinically complex than other MET patients. Experiencing a MET call away from the usual-care unit in the hospital in an area such as the RD may place patients at increased risk of death and in need of escalation of care. Further research in this area is needed. Transient care of patients outside of their usual care areas in the RD potentially places patients at risk. Better mechanisms for identifying at-risk patients may improve the utilisation of resources for nursing surveillance in the RD, thus improving patient outcomes. Improved communication and handover between the RD and the nursing staff from the referring units could increase the RD nurses' ability to identify patient care needs, appropriate monitoring and surveillance. In the RD, patients clearly represent a special clientele for METs and RRSs and merit further study.
We are grateful to Marlon Johnson, Nikhil Amesur, Melissa Saul, Anu Pillai and the nurses and technicians of the UPMC Presbyterian Shadyside Imaging Services for their support of our work and their commitment to improving patient care.
Linked article 000817
Funding Lora Ott is supported by NINR National Research Service Award (1F31NR012343-01) and the Leslie A. Hoffman Endowed Research Award 2010, Michael R Pinsky, MD by NHLBI mid-career development award (2K24 HL067181) and Sunday Clark is supported by NCRR NIH Roadmap for Medical Research (KL2 RR024154).
Competing interests None.
Ethics approval Ethics approval was provided by Institutional Review Board University of Pittsburgh.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement The data obtained in this study are the property of the principal researcher, the first named author.