Utilization and outcomes of tracheostomies in the intensive care unit in Iceland in 2007-2020: A descriptive study

BACKGROUND

Tracheostomies are commonly utilized in ICU patients due to prolonged mechanical ventilation, upper airway obstruction, or surgery in the face/neck region. However, practices regarding the timing of placement and utilization vary. This study provides a nationwide overview of tracheostomy utilization and outcomes in the ICU over a 14-year period.

METHODS

A retrospective study including all patients that received a tracheostomy during their ICU stay in Iceland between 2007 and 2020. Data were retrieved from hospital records on admission cause, comorbidities, indication for tracheostomy insertion, duration of mechanical ventilation before and after tracheostomy placement, extubation attempts, complications, length of ICU and hospital stay and survival. Descriptive statistics were provided, and survival analysis was performed using Cox regression.

RESULTS

A total of 336 patients (median age 64 years, 33% females) received a tracheostomy during the study period. The most common indication for tracheostomy insertion was respiratory failure, followed by neurological disorders. The median duration of mechanical ventilation prior to tracheostomy insertion was 9 days and at least one extubation had been attempted in 35% of the cases. Percutaneous tracheostomies were 32%. The overall rate of complications was 25% and the most common short-term complication was bleeding (5%). In-hospital mortality was 33%. The one- and five-year survival rate was 60% and 44%, respectively.

CONCLUSIONS

We describe a whole-nation practice of tracheostomies. A notable finding is the relatively low rate of extubation attempts prior to tracheostomy insertion. Future work should focus on standardization of assessing the need for tracheostomy and the role of extubation attempts prior to tracheostomy placement.

Management of acute atrial fibrillation in the intensive care unit: An international survey

BACKGROUND

Atrial fibrillation (AF) is common in intensive care unit (ICU) patients and is associated with poor outcomes. Different management strategies exist, but the evidence is limited and derived from non-ICU patients. This international survey of ICU doctors evaluated the preferred management of acute AF in ICU patients.

METHOD

We conducted an international online survey of ICU doctors with 27 questions about the preferred management of acute AF in the ICU, including antiarrhythmic therapy in hemodynamically stable and unstable patients and use of anticoagulant therapy.

RESULTS

A total of 910 respondents from 70 ICUs in 14 countries participated in the survey with 24%-100% of doctors from sites responding. Most ICUs (80%) did not have a local guideline for the management of acute AF. The preferred first-line strategy for the management of hemodynamically stable patients with acute AF was observation (95% of respondents), rhythm control (3%), or rate control (2%). For hemodynamically unstable patients, the preferred strategy was observation (48%), rhythm control (48%), or rate control (4%). Overall, preferred antiarrhythmic interventions included amiodarone, direct current cardioversion, beta-blockers other than sotalol, and magnesium in that order. A total of 67% preferred using anticoagulant therapy in ICU patients with AF, among whom 61% preferred therapeutic dose anticoagulants and 39% prophylactic dose anticoagulants.

CONCLUSION

This international survey indicated considerable practice variation among ICU doctors in the clinical management of acute AF, including the overall management strategies and the use of antiarrhythmic interventions and anticoagulants.

Predictive variables for poor long-term physical recovery after intensive care unit stay: An exploratory study

BACKGROUND

Elucidating factors that influence physical recovery of survivors after an intensive care unit (ICU) stay is paramount in maximizing long-term functional outcomes. We examined potential predictors for poor long-term physical recovery in ICU survivors.

METHODS

Based on secondary analysis of a trial of 50 ICU patients who underwent mobilization in the ICU and were followed for one year, linear regression analysis examined the associations of exposure variables (baseline characteristics, severity of illness variables, ICU-related variables, and lengths of ICU and hospital stay), with physical recovery variables (muscle strength, exercise capacity, and self-reported physical function), measured one year after ICU discharge.

RESULTS

When the data were adjusted for age, female gender was associated with reduced muscle strength (P = .003), exercise capacity (P < .0001), and self-reported physical function (P = .01). Older age, when adjusted for gender, was associated with reduced exercise capacity (P < .001). After adjusting for gender and age, an association was observed between a lower score on one or two physical recovery variables and exposure variables, specifically, high body mass index, low functional independence, comorbidity and low self-reported physical function at baseline, muscle weakness at ICU discharge, and longer hospital stay. No adjustment was made for cumulative type I error rate due to small number of participants.

CONCLUSION

Elucidating risk factors for poor long-term physical recovery after ICU stay, including gender, may be critical if mobilization and exercise are to be prescribed expediently during and after ICU stay, to ensure maximal long-term recovery.

The potential for organ donation in Iceland: A nationwide study of deaths in intensive care units

BACKGROUND

The deceased organ donation rate in Iceland has been low compared with other Western countries. The aim of this study was to explore the potential for organ donation after brain death in Iceland.

METHODS

Observational cohort study of patients with catastrophic brain injury who died in intensive care units (ICUs) at hospitals in Iceland in 2003-2016. Medical records were retrospectively reviewed to identify potential donors (PDs), using the WHO Critical Pathway for Deceased Donation. Trends in annual incidence of PDs, conversion to actual donors, and family refusals were assessed.

RESULTS

Among 1537 patients who died in the ICU, 125 (8.1%) were identified as PDs. Of 103 PDs who were declared brain dead, consent for organ donation was pursued in 84 cases and granted in 63. Fifty-six became actual donors. The annual donation rate averaged 13 per million population (pmp), but rose abruptly in the final 2 years to 36 and 27 pmp, respectively. This was paralleled by an increase in annual incidence of PDs from an average of 28 pmp to 54 and 42 pmp, respectively. The donor conversion rate increased during the study period (P = .026). Twenty-three PDs (18%) were not pursued without an apparent reason.

CONCLUSIONS

The donation rate increased markedly in the last 2 years of the study period after remaining low for more than a decade. This change can largely be explained by a high incidence of PDs and a low family refusal rate. Missed donation opportunities suggest a potential to maintain a high donation rate in the future.

Long-term survival after intensive care: A retrospective cohort study

Background Limited data exist on long-term survival of patients requiring admission to intensive care units (ICUs). The aim of this study was to investigate long-term survival of ICU patients in Iceland and assess changes over a 15-year period. Methods Data were collected on age, gender, admission cause, length of stay, comorbidities, mechanical ventilation and survival of patients 18 years and older admitted to the ICUs in Landspitali during 2002-2016. Long-term survival of patients surviving more than 30 days from admission was estimated and its predictors assessed with Cox regression analysis. Long-term survival was compared to the survival of an age- and gender-matched reference group from the general population. Results Of 15 832 ICU admissions, 55% was medical, 38% was surgical and 7% was due to trauma. The 5-year survival of medical, surgical and trauma patients was 66%, 76% and 92% respectively. Significant survival differences were found between admission subgroups. Higher age and comorbidity burden was related to decreased survival in all patient groups. After correcting for age, gender, comorbidities, length of ICU stay and mechanical ventilation, patient survival improved during the study period only for patients admitted for infections. There was a high variability in the estimated time point where the ICU admission had no residual effect on survival. Conclusions Long-term survival of ICU patients is substantially decreased compared to the general population, but varies based on admission causes. Improved long-term survival of patients admitted with infections could be explained by earlier detection and improved treatment of septic shock.

A population-based study on epidemiology of intensive care unit treated traumatic brain injury in Iceland

BACKGROUND

Traumatic brain injury is a worldwide health issue and a significant cause of preventable deaths and disabilities. We aimed to describe population-based data on intensive care treated traumatic brain injury in Iceland over 15 years period.

METHODS

Retrospective review of all intensive care unit admissions due to traumatic brain injury at The National University Hospital of Iceland 1999-2013. Data were collected on demographics, mechanism of injury, alcohol consumption, glasgow come scale upon admission, Injury Severity Scoring, acute physiology and chronic health evaluation II score, length of stay, interventions and mortality (defined as glasgow outcome score one). All computerized tomography scans were reviewed for Marshall score classification.

RESULTS

Intensive care unit admissions due to traumatic brain injury were 583. The incidence decreased significantly from 14/100.000/year to 12/100.000/year. Males were 72% and the mean age was 41 year. Majority of patients (42%) had severe traumatic brain injury. The most common mechanism of injury was a fall from low heights (36.3%). The mortality was 18.2%. Increasing age, injury severity score, Marshall score and acute physiology and chronic health evaluation II score are all independent risk factors for death. Glasgow coma scale was not an independent prognostic factor for outcome.

CONCLUSIONS

Incidence decreased with a shift in injury mechanism from road traffic accidents to falls and an increased rate of traumatic brain injury in older patients following a fall from standing or low heights. Mortality was higher in older patients falling from low heights than in younger patients suffering multiple injuries in road traffic accidents. Age, injury severity score, acute physiology and chronic health evaluation II score and Marshall score are good prognostic factors for outcome. Traumatic brain injury continues to be a considerable problem and the increase in severe traumatic brain injury in the middle age and older age groups after a seemingly innocent accident needs a special attention.

Evaluation of clinical efficacy and safety of cervical trauma collars: differences in immobilization, effect on jugular venous pressure and patient comfort

BACKGROUND

Concern has been raised that cervical collars may increase intracranial pressure in traumatic brain injury. The purpose of this study was to compare four types of cervical collars regarding efficacy of immobilizing the neck, effect on jugular venous pressure (JVP), as a surrogate for possible effect on intracranial pressure, and patient comfort in healthy volunteers.

METHODS

The characteristics of four widely used cervical collars (Laerdal Stifneck(®) (SN), Vista(®) (VI), Miami J Advanced(®) (MJ), Philadelphia(®) (PH)) were studied in ten volunteers. Neck movement was measured with goniometry, JVP was measured directly through an endovascular catheter and participants graded the collars according to comfort on a scale 1-5.

RESULTS

The mean age of participants was 27 ± 5 yr and BMI 26 ± 5. The mean neck movement (53 ± 9°) decreased significantly with all the collars (p < 0.001) from 18 ± 7° to 25 ± 9° (SN < MJ < PH < VI). There was a significant increase in mean JVP (9.4 ± 1.4 mmHg) with three of the collars, but not with SN, from 10.5 ± 2.1 mmHg to 16.3 ± 3.3 mmHg (SN < MJ < VI < PH). The grade of comfort between collars varied from 4.2 ± 0.8 to 2.2 ± 0.8 (VI > MJ > SN > PH).

CONCLUSION

Stifneck and Miami J collars offered the most efficient immobilization of the neck with the least effect on JVP. Vista and Miami J were the most comfortable ones. The methodology used in this study may offer a new approach to evaluate clinical efficacy and safety of neck collars and aid their continued development.

Induced hypothermia in comatose survivors of asphyxia: a case series of 14 consecutive cases

BACKGROUND

Induced hypothermia is widely used for comatose survivors of cardiac arrest. Other causes of hypoxic brain injury carry a poor prognosis when treated using traditional methods. At our hospital, hypothermia has also been used for the management of all comatose survivors of asphyxiation. The aim of the present study was to report the results of the management of these patients.

METHODS

Hospital charts of all patients admitted unconscious after asphyxiation during a 7-year period were reviewed. This included patients after hanging, drowning, carbon monoxide intoxication and other gas intoxications. In all patients, hypothermia with a target temperature of 32-34 degrees C was induced with external or intravascular cooling for 24 h. The primary outcome was neurologic function at discharge.

RESULTS

Fourteen male patients were treated with hypothermia, eight after hanging, three after drowning, two after carbon monoxide intoxication and one after methane intoxication. All were deeply comatose (Glasgow Coma Score 3-5) on arrival to hospital. Nine had been resuscitated from cardiac arrest. There were nine survivors (65%), all with good neurological recovery (Cerebral Performance Category 1-2). Four out of five non-survivors showed cerebral edema already on arrival computed tomographic (CT) scan while none of the nine survivors did.

CONCLUSIONS

The results of this study suggest that an early abnormal CT scan of the brain in patients resuscitated after asphyxiation carries an adverse prognosis. The favorable outcome of the patients in the present study suggests that a randomized clinical trial on the use of induced hypothermia in patients exposed to severe asphyxia might be warranted.

[Energy expenditure and nutritional support in intensive care patients]

STUDY OBJECTIVES

Nutritional support of ICU patients is usually guided by estimations of their caloric needs. However, recent studies have shown that energy expenditure (EE) of critically ill patients is not as high as previously thought. The goal of this study was to measure EE in ICU patients, compare it with estimated EE and evaluate nutritional support.

METHODS

Energy expenditure was measured with indirect calorimetry in a broad group of ICU patients requiring mechanical ventilation >48 hours. In comparison EE was estimated with the Harris-Benedict equation. Nutritional support during ICU stay was registered.

RESULTS

Mean measured EE of 56 patients was 1820 +/- 419 kcal/day or 22 kcal/kg/day. The Harris- Benedict equation underestimated EE by 11%, but adding a stress factor resulted in 15% overestimation. Mean nutritional support was 1175 +/- 442 kcal/day or 67% of EE. The energy deficit was greatest during the first week of ICU stay. Mean protein administration was 0,44 g/kg/day.

CONCLUSION

Measured EE of ICU patients was less than nutritional support recommended by international guidelines. These results are in accordance with recent studies. Nutritional support was only 67% of measured energy expenditure and protein content less than recommended. Further studies are needed as it has not be shown how this might influence outcome.

[The incidence and mortality of ARDS in Icelandic intensive care units 1988-1997]

OBJECTIVE

A retrospective analysis of the epidemiology and intensive care treatment of ARDS in Iceland during the 10 year period, 1988-1997 with observation of trends within the period.

MATERIAL AND METHODS

All ICU admissions in Iceland 1988-1997 were reviewed according to the American-European consensus conference criteria on ARDS to select patients with the diagnosis of ARDS i.e. bilateral pulmonary infiltrates, PaO(2)/FiO(2) <200 and excluding patients with signs of heart failure or a pulmonary capillary wedge pressure (PCWP) >18 mmHg. Data were collected on age, gender, length of stay, ventilator treatment and ventilatory modes, causes of ARDS and mortality.

RESULTS

A total of 220 patients with severe respiratory failure were found and 155 of them were diagnosed as having ARDS or an annual incidence of 15.5 cases/year or 5.9 cases/100.000/year. If reference population >15 years of age is used for calculation the incidence is 7.8 cases/100.000/year. Hospital mortality was 40%, mean length of ICU stay was 21 days, mean hospital length of stay 39 days. The incidence of ARDS increased during the period with a tendency to lower mortality rates. Mortality was significantly lower when pressure controlled ventilation was used, compared to volume controlled ventilation.

CONCLUSION

The incidence of ARDS in a well defined population of Iceland is lower than recent studies in USA and Europe have shown or 5.9 cases/100.00/year but is increasing. The mortality is 40% and shows a slight downward trend, which may be due to the use of lung protective ventilation.

[Organ donations in Iceland 1992-2002]

OBJECTIVE

To acquire information of organ donations and organ waiting lists in Iceland 1992-2002, the beginning of an organ procurement system.

MATERIAL AND METHODS

Records of all patients treated at the ICU at LSH in Fossvogur 1992-2002 were studied. Information of organ donations at other units and number on organ waiting lists was attained. Results are shown as medians (25th, 75th percentile).

RESULTS

527 patients died at the ward 1992-2002 (48 (45, 52) annually). Of them 68 (13%) were declared deceased because of brain death. Permission for organ procurement was requested from relatives in 50 cases (74% of brain deaths) and was obtained in 30 (60%). Four of these were not suitable as organ donors. Organ donation was denied in 40% of cases and refusal seemed to increase during the period. Potential organ donors that were not recognised as such were 18A (3% of all deaths). During the period six organ donations occurred at others units. The total number of organ donors was 32 (3 (1, 5) annually) in Iceland 1992-2002. Number of organs donated was 109 (11 (4, 15) annually). Each year there were 7 (5, 9) individuals on waiting lists for organs and 3 (2, 5) were transplanted.

CONCLUSIONS

87% of organ donations in Iceland come from patients with cerebral haemorrhage, traumatic brain injury or ischaemic stroke. Organ donations in Iceland seem to fulfill the nations need of organs. It is a possible worry that refusal of organ donation by relatives seemed to increase during the period.

[Neuromuscular monitoring during anesthesia.]

OBJECTIVE

Muscle relaxants are very important in anesthetic practice but must be used with great care. Studies have shown that 17-40% of patients in postanesthesia care units (PACU) have residual muscle weakness. The purpose of this study was to evaluate whether the use of neuromuscular monitors during anesthesia could reduce the incidence of muscle weakness in the postoperative period.

MATERIALS AND METHOD

Eighty patients operated for laparoscopic cholecystectomy or lumbal disc prolapse given muscle relaxants during anesthesia were studied, randomly allocated to four groups. Fourty of these patients were monitored with neuromuscular monitor (TOF-guard") during anesthesia and the set point was a TOF-ratio of at least 70% before extubating the patients. Fourty patients were monitored by usual clinical signs (spontaneus breathing, cough and muscle movement). Twenty patients in each group were given vecuronium as muscle relaxant and 20 patients recieved pancuronium, again patients were randomly selected. In the PACU all patients were evaluated and the "5-sec headlift test" was used to find patients with muscle weakness. Hand grip strength was also measured before anesthesia and in the PACU. Glascow Coma Score (GCS) was used to evaluate if patients were too drowsy to co-operate and patients with GCS < 12 were excluded. Measurements were made after arrival to the PACU and every 30 minutes thereafter until headlift was at least 5 sec.

RESULTS

The incidence of restcurarization was 15% on arrival to the PACU. No statistically significant difference was found between those monitored with neuromuscular monitors and those that were not. Similarily no statistical difference was found between short acting neuromuscular blocking agents and longer acting agents.

CONCLUSION

Although we didn t find any benefit from neuromuscular monitoring or using shorter acting drugs, the use of nervestimulators and short acting drugs is still recommended, especially for high risk patients. The generally accepted train-of-four (TOF-) ratio of 70% has been questioned by some authors, recommending a higher ratio (85%). Further studies using a higher TOF-ratio are therefore recommended.

Severe atenolol and diltiazem overdose

CASE REPORT

A case of combined, massive overdose of both atenolol and diltiazem in an adult male is reported. Cardiac arrest ensued which was responsive to cardiopulmonary resuscitation. Bradycardia, hypotension, and oliguria followed which were resistant to intravenous pacing and multiple pharmacologic interventions, including intravenous fluids, calcium, dopamine, dobutamine, epinephrine, prenalterol, and glucagon. Adequate mean arterial pressure and urine output were restored only after addition of phenylephrine to therapy with multiple agents and transvenous pacing. The patient survived until discharge after a hospital course complicated by nontransmural myocardial infarct on hospital day 4 and pneumonia. Laboratory testing subsequently revealed high serum levels of both atenolol and diltiazem. The atenolol level of 35 microg/mL in this patient is the highest reported associated with survival.

CONCLUSION

This case illustrates severe cardiovascular toxicity after overdose of both atenolol and diltiazem. Oliguria, which has previously been reported in severe atenolol overdose, was successfully treated without hemodialysis by the addition of phenylephrine to aggressive therapy with pacing, inotropic, and pressor support.

[Thirty years of intensive care. Clinical experience at Reykjavik Hospital.]

OBJECTIVE

Reykjavik Hospital has been the main trauma center in Iceland. The Intensive Care Unit (ICU) was founded in 1970 and has been in operation since then. The aim of this study was to review its clinical experience these 30 years.

MATERIAL AND METHODS

A retrospective study of patient records was conducted for all admissions to the ICU between 1970 and the end of 1999. Data was collected pertaining to the annual rate of admission, proportion of patients requiring ventilator treatment, mortality rate, age distribution, reasons for admission and medical speciality.

RESULTS

A total of 13,154 patients were admitted to the ICU between 1970 and the end of 1999. A steady increase in the rate of admissions was observed during the study period, reaching 550-600 patients for the ICU annually. There was a statistically significant increase in the proportion of patients requiring ventilator treatment over the study period, reaching 38% of ICU admissions by the end of the study. During the study period only one statistically significant change was observed in age distribution. The annual rate of admission to the ICU for patients over 60 years of age increased significantly between the periods 1985-1989 and 1990-1999. The proportion of surgical patients increased (70% of patients by the end of the study) and the proportion of medical patients decreased (ending at 30% of patients). During the last decade a significant increase was seen in patients admitted after major surgery. The observed mortality rate in the final years of the study was observed to be significantly less than it had been in previous years. The observed mortality rate from 1970 to 1989 was 11.7% of patients, decreasing to 8.6% from 1990 to 1998. The average length of stay was also observed to decline.

CONCLUSIONS

The decline in mortality occurred in spite of an increased rate of admission and an increased workload. This change is attributed to improvement in the care of critically ill patients over the study period.

[Head injury at Reykjavík Hospital, intensive care unit, 1994-1998.]

OBJECTIVE

Reykjavík Hospital is the main trauma hospital in Iceland, receiving all severe head injuries in the country. Incidence of head injury and mortality has been decreasing in the last decades. The aim of this study was to analyse data on admission, treatment and outcome of patients admitted to intensive care unit with severe head injury and compare with other countries.

MATERIAL AND METHODS

In this study we looked retrospectively at the incidence of severe head injuries admitted to the intensive care unit at Reykjavik Hospital 1994-1998. Number of patients, type of injury, length of stay, length of ventilator treatment. Glasgow Coma Score (GCS), APACHE II (Acute Physiologic and Chronic Health Evaluation) score and mortality was analysed.

RESULTS

A total of 236 patients was admitted with an average of 47 patients per year. Traffic accidents were the most common cause of injury and mortality was 11.7%. Ethanol consumption was seen in many cases where fall was the cause of accident, most often in the year 1998 in 75% of cases. Mortality of patients with GCS 8 that was 40% of the patients was must higher or 24.7% compared with patients with GCS >8 where mortality was 3.4%. There was an increase in admissions in 1998, with more severe injuries and significantly longer length of stay and ventilator treatment.

CONCLUSIONS

Number of patients with head injury was decreasing in comparison with older studies. The results of treatment are rather good in comparison with other countries with relatively low mortality, or 11.7% versus 15-20% in nearby countries. There has been improvement of outcome in patients with the most severe head injury (GCS 8) since 20 years ago, where up to 50% of the patients died but in our study mortality was 24.7%. Alcohol consumption was seen in 46% of cases where fall was the cause of head injury. Those that suffer head trauma are most often young people and preventive measures must continue with full strength in order to decrease the incidence of accidents in our society.

[Postoperative restcurarization.]

OBJECTIVE

Muscle relaxants have been used during anesthesia for the past 50 years but in the last decades it has been realised that their use can lead to complications. Studies have shown 20-40% incidence of restcurarization in postanesthesia care units (PACU) even if neuromuscular monitors are used during anesthesia. The purpose of this study was to estimate the frequency of postoperative muscle weakness at the Reykjavik Hospital in Iceland.

MATERIAL AND METHODS

Sixty patients, operated for laparoscopic cholecystectomy or lumbal disc prolapse, given muscle relaxants (vecuronium or pancu notronium) during anesthesia were studied in the PACU. The 5-sec headlift test has been shown to be the best clinical sign of recovery and this sign was used to find patients with muscle weakness. Glasgow coma score (GCS) was used to evaluate if patients were too drowsy to co-operate and patients with GCS <12 were excluded. Measurements were made after arrival to the PACU and every 30 minutes thereafter until headlift was at least five seconds.

RESULTS

Incidence of restcurarization was 17% on arrival to the PACU and these patients were significantly lower in oxygen saturation before oxygen supplementation was started. Thirty minutes after arrival 6% were still restcurarized and 3% after 60 minutes. All patients had recovered after 90 minutes. No difference was found between patients given vecuronium or pancuronium in the first two measurements but those with longest duration of muscle weakness had received pancuronium.

CONCLUSION

The study shows that the incidence of muscle weakness is too high, which might increase the risk for complications such as hypoxia or respiratory failure. To increase patient safety, shorter acting drugs are recommended and the use of new nervestimulators giving the train-of-four(TOF)-ratio during muscle blockade could possibly improve the situation.