1. Chu, Julie J. MSN, CRNP
  2. Walters-Fischer, Patricia BS, RN

Article Content


Reviewing the evidence to optimize care.

A review of the evidence on protocols of spinal immobilization, complications resulting from the intervention, and efficacy of several imaging modalities has shown that when prolonged, spinal immobilization often results in complications such as pressure ulcers. If cervical injury in patients who are unconscious with multiple injuries is ruled out by plain radiography and directed computed tomography, they shouldn't continue to be immobilized even if they aren't expected to be conscious within 48 to 72 hours.


Researchers reviewed several databases (MEDLINE [1960 to 2004], EMBASE, PubMed, and the Cochrane Library), contacted clinicians, and examined Web sites for literature containing the subjects "cervical spine and cord injury," "cervical collar," "spinal precautions and immobility," "plain radiography," and "computed tomography," among others.


They found that the development of pressure ulcers from the prolonged use of cervical collars (49 to 72 hours) is a common occurrence (in as many as 55% of patients), and other complications included elevated intracranial pressure and venous obstruction loss of airway, failed enteral nutrition, gastroesophageal reflux, pulmonary aspiration, and thromboembolism. The review determined also that because plain radiography of the cervical spine and directed computed tomography are as effective as magnetic resonance imaging and dynamic fluoroscopy in detecting an unstable cervical spinal injury (more than 99% of such injuries are thereby detectable) the former procedures should be used as soon as possible to rule out injury and to allow for supervised patient mobilization, even if the patient remains unconscious.-JC


Morris CG, et al. BMJ 2004;329(7464):495-9.



It doesn't reduce readmission rates among patients with heart failure.

A randomized study has revealed that customizing education in heart failure and its management according to patients' "health beliefs" doesn't decrease readmission rates associated with heart failure or necessarily improve the quality of their lives, compared with patients who don't receive that intervention. But it does alter their perceptions of the benefits and drawbacks of heart failure management.


Using the Health Belief Model as a framework, 70 patients with heart failure at a community hospital were randomly assigned to receive either an individualized educative intervention during hospitalization and both one week and one month after discharge (intervention group, n = 33) or usual discharge instructions (control group, n = 37). The primary outcome studied was hospital readmission rate, assessed three months after discharge, and the quality of life in all patients was measured with the Minnesota Living with Heart Failure Questionnaire at baseline and one month after discharge. Patients' perceptions of the benefits and drawbacks of taking heart failure medications, following a sodium-restricted diet, and self-monitoring for signs of fluid overload were measured using components of the Health Belief Scales at baseline, one week, and one month after discharge in the treatment group only; individualized education was provided accordingly. Patients in the control group received follow-up telephone calls to assess the quality of life after one month.


The results showed that readmission rates weren't significantly different between the two groups after three months, nor were quality-of-life scores, according to univariate analyses. But the customized educative intervention was associated with favorable changes in patients' perceptions of the benefits and drawbacks of heart failure management. According to univariate analyses, significant changes occurred between baseline and one week, and between baseline and one month, in the scores pertaining to drawbacks of medication use, benefits of diet, drawbacks of diet, benefits of self-monitoring, and drawbacks of self-monitoring. The benefits-of-medication-use scores didn't change significantly.--JC


Sethares KA, Elliott K. Heart Lung 2004; 33(4):249-60.



Favorable outcomes shown with a nurse-delivered protocol.

A study conducted in the United Kingdom indicates that increasing or maintaining the stroke index above 35 mL/m2 after a patient undergoes cardiac surgery can reduce the incidence of complications and shorten both the ICU and the total hospital lengths of stay.


Between April 2000 and January 2003, 179 patients were recruited and randomly assigned to either a protocol group or a control group. Nurses followed the protocol to optimize circulatory status in patients immediately after cardiac surgery. In patients in both groups, the nurse inserted an esophageal Doppler probe within 10 minutes of arrival at the ICU after cardiopulmonary bypass surgery. The probe detected the descending aortic blood flow, displayed on the monitor by a waveform, the characteristics of which (size and shape) were observed by the nurse, enabling her to effectively monitor left ventricular contractibility, preload, and afterload. Patients were evaluated again after four hours.


In the protocol group, an algorithm was followed to increase the stroke volume index to 35 mL/m2 through the use of IV fluids and also inotropes and nitrates, as necessary. The index in the control group was not maintained at 35 mL/m2, those patients being treated according to standard hospital protocol and reevaluated at four hours.


Because complications were identified more easily and addressed more quickly in the protocol group, the result was a decrease in duration of hospital stay from a median of nine days to seven, while the mean duration of hospital stay decreased from 13.9 days to 11.4.--PWF


McKendry M, et al. BMJ 2004;329(7460): 258.



The quality of end-of-life care may be reflected in nurses' assessments.

A prospective cohort study has demonstrated the feasibility of using a questionnaire to measure the quality of dying and death in the ICU. The absence of cardiopulmonary resuscitation during the last eight hours of life, the presence of someone at the time of death, and care managed by neurosurgery or neurology services were the three factors associated with higher scores.


A total of 178 patients who died during an eight-month period in the ICU at the level I trauma center at Harborview Medical Center, University of Washington, Seattle, were studied using a nursing assessment tool to rate the quality of dying and death. The nurses rated the quality of each of 14 items on a scale of 0 ("terrible experience") to 10 ("almost perfect experience"). Other demographic and clinical data were collected from medical records.


The two most common diagnoses of the patients, among whom the mean age was 54 years, were trauma and intracranial hemorrhage. The response rate for the questionnaires was quite high (72.4%). The mean total score from 149 questionnaires was 73.1. Multivariate analysis revealed that the absence of cardiopulmonary resuscitation during the last eight hours of life, the presence of someone at the time of death, and care managed by neurosurgery or neurology services were independent predictors of the quality-of-dying-and-death score.


The results of the study suggest that improvement in the quality of end-of-life care could be attained by focus on the three factors found to be associated with higher scores, but its limitations include the inability to generalize the results to other institutions and the use of only nurses' perspectives in the assessment of the quality of dying and death.--JC


Hodde NM, et al. Crit Care Med 2004; 32(8):1648-53.



Improving the end-of-life care of patients with dementia.

A care management approach in which an inpatient palliative care team intervenes in the care of critically ill patients with terminal dementia improves the quality of end-of-life care and diminishes the unnecessary use of resources, according to a recently published study.


In a two-phase study, 26 patients with advanced dementia (present before hospitalization) at the Detroit Receiving Hospital medical ICU (MICU) in Michigan were selected from a retrospective chart review during 1998 to 1999 as a historical control group. Twenty-six comparable patients were prospectively enrolled during the second phase from 1999 to 2001, when anticipatory palliative care-including communicating patients' prognoses to family members; identifying patients' advance directives or end-of-life care wishes; discussing treatment options with family members or surrogates; implementing "comfort measures only," when applicable; and educating the health care team-was instituted. Data such as age, sex, Acute Physiology and Chronic Health Evaluation score, and Therapeutic Intervention Scoring System points were collected.


The results showed that patients in the intervention group had significantly shorter hospital and MICU mean lengths of stay, compared with the control group, 3.5 and 6.8 days, respectively, and 7.4 and 12.1 days, respectively. Although 73% of patients in the control group had do-not-resuscitate orders, all patients in the intervention group had them. In addition, there were significantly more cases in which goals reflected "comfort measures only" (62%) in the intervention group, compared with the control group (16%).


A palliative care team in the ICU can significantly improve the end-of-life care of critically ill patients with terminal dementia, and ICU nurses can contribute to those goals by working collaboratively and bearing in mind the necessity of performing palliative care if treatment goals change.--JC


Campbell ML, Guzman JA. Crit Care Med 2004;32(9):1839-43.



Interventional strategies are compared.

A recently published study shows that treating patients who have ST-segment elevation myocardial infarction (STEMI) with cardiac catheterization and revascularization (if appropriate) soon after thrombolysis is associated with better outcomes than those of an ischemia-guided, conservative approach.


In a randomized, multicenter trial conducted in Spain between 2000 and 2001, 500 patients with STEMI were assigned either to an interventional-strategy group, in which patients underwent coronary angiography and revascularization, when appropriate, within 24 hours of thrombolysis with recombinant tissue plasminogen activator (n = 248) or to an ischemia-guided, conservative-strategy group, in which the same interventions were performed only in cases of spontaneous or stress-induced ischemia (n = 252). All patients received standard postmyocardial medication, but patients in the interventional-strategy group also received the antiplatelet agents ticlopidine or clopidogrel, as well as abciximab. The primary end point was the combined rate of mortality, nonfatal reinfarction, or ischemia-induced revascularization at one year.


The largest proportion of patients in the interventional-strategy group received stenting of the "culprit" artery (80%)-followed by medical treatment only (17%), coronary bypass surgery (2%), and stenting of a "nonculprit" artery (1%). In the conservative-strategy group, 20% of patients received ischemia-guided revascularization with stenting before discharge. The incidence of revascularization attributable to spontaneous ischemia was nearly five times higher in the conservative-strategy group during the index hospital stay. The rates of complications, such as significant bleeding, were low and comparable in the two groups. The difference in the mean index hospital stay was approximately three days shorter in the interventional-strategy group. At the 30-day point, mortality rates and reinfarction rates were comparable in both groups, but at the one-year point, significantly more patients in the conservative-strategy group had reached the primary combined end point, compared with the interventional-strategy group: 21% versus 9%, respectively.--JC


Fernandez-Aviles F, et al. Lancet 2004; 364(9439):1045-53.



Two assessment tools are compared.

The categories alert-verbal-painful-unresponsive in the AVPU responsiveness scale appear to correspond to ranges of scores in the Glasgow Coma Scale (GCS) and may be easier to use than the latter, according to a recently published study.


Using the two assessment tools concurrently, nurses assessed the consciousness levels of 1,384 patients admitted with either intentional or accidental poisoning to the toxicology unit of the ED of an academic teaching hospital during 2002 and 2003. Assessment according to the AVPU responsiveness scale entailed the use of an algorithm in which patients who opened their eyes spontaneously, demonstrated orientation in speech, and obeyed commands were deemed "alert." If there was any verbal, motor, or ocular response to a verbal stimulus or to a gentle shake in the event that the patient was either hard of hearing or asleep, patients were deemed "responsive to verbal stimulation." Patients were considered "responsive to painful stimulation" upon any verbal, motor, or ocular response to a painful stimulus. If there was no response, patients were considered "unresponsive." The median GCS scores and their ranges were compared with the interquartile ranges (IQRs) in each AVPU responsiveness category.


The majority of the patients was found to be "alert" (82%), 10% were responsive to verbal stimulus, 6% were responsive to painful stimulus, and 1% were unresponsive. There was some overlapping of the ranges of the two assessment tools. For example, the median GCS score in the "alert" category was 15 (GCS range, 8 to 15; IQR, 15), 13 in the "responsive to verbal stimulation" category (GCS range, 7 to 14; IQR, 12 to 14), 8 in the "responsive to painful stimulation" category (GCS range, 5 to 13; IQR, 7 to 9), and 3 in the "unresponsive" category (GCS range, 3 to 6; IQR, 3). All patients with an AVPU responsiveness scale score within the "unresponsive" category also had a GCS score lower than or equal to 6 and needed intubation. Regardless of the assessment tool used, nurses considered the patients who were intoxicated with alcohol to be the most difficult to assess. They also rated the AVPU responsiveness scale as easier to use than the GCS. -JC


Kelly CA, et al. Ann Emerg Med 2004;44(2): 108-13.



The optimal 'demand' dose is determined.

A recent study has revealed that the administration of fentanyl by patient-controlled analgesia (PCA) provides pain management during burn dressing changes most effectively when an initial loading dose of IV 1 mcg/kg is administered and followed by a "demand" dose of 30 mcg with a five-minute lockout interval.


Sixty patients with thermal burns on more than 20% of body surface area (in otherwise normal health or with mild systemic disease) who were scheduled to receive burn dressing changes were randomly assigned to receive a demand dose of fentanyl 10 mcg (n = 15), 20 mcg (n = 15), 30 mcg (n = 15), or 40 mcg (n = 15) by PCA (with a five-minute lockout interval) after an initial loading dose of IV fentanyl 1 mcg/kg 10 minutes prior to the dressing change. The patients were instructed to press the PCA demand dose button during the change of dressing when they felt that pain intensity became greater than 2 on the visual analog scale (VAS). Hemodynamic and respiratory variables were recorded every 10 minutes, and level of sedation also was assessed. Postprocedural measurements were recorded every 15 minutes for two hours.


Patients who received 10 mcg or 20 mcg of fentanyl had significantly higher mean VAS scores compared with patients who received 30 mcg or 40 mcg of it (7.73 +/- 1.33 and 7.20 +/- 1.21 versus 4.47 +/- 0.83 and 3.90 +/- 0.63, respectively), correlating with the greater number of dose demands in the 10-mg and 20-mg groups. The analgesic efficacy in the 30 mcg and 40 mcg groups was comparable. Fentanyl delivery by PCA was tolerated well, and there were no significant deviations from baseline hemodynamic or respiratory status in any patient. -JC


Prakash S, et al. Anesth Analg 2004;99(2): 552-5.



A study examines the risk factors.

Patients in the pediatric intensive care unit (PICU) who undergo many invasive procedures are at greater risk of suffering psychological sequelae six weeks after discharge than are those who aren't exposed to as many, according to a recent study.


Using the data of a previous prospective cohort study of the psychological responses of children after critical illness, researchers analyzed that pertaining to 60 six-to-17-year-old children hospitalized in the PICUs of two Canadian pediatric hospitals to identify patients who were at high risk of experiencing psychological sequelae after discharge. Patients were categorized into high-risk and low-risk groups according to severity of illness, as measured by the Pediatric Risk of Mortality Scale, and to the number of invasive procedures that they were exposed to, as measured by the Invasive Procedure Scores. After both six weeks and six months of discharge, patients completed the Children's Impact of Events Scale, the Children's Medical Fears Scale, and the Children's Health Locus of Control Scale questionnaires.


Statistically and clinically significant differences were found six weeks after discharge. Patients in the high-risk group reported more medical fears, perceived that they had less control over their health, and reported more posttraumatic stress symptoms, compared with patients in the low-risk group, and the symptoms continued to be greater in the high-risk group six months after discharge. According to multiple linear regression analyses, the number of invasive procedures to which patients were exposed was found to have the greatest effect on the observed differences between the groups at six weeks after discharge, whereas the severity of illness wasn't a significant factor. -JC


Rennick JE, et al. Pediatr Crit Care Med 2004;5(4):358-63.