Proceedings of Singapore Healthcare

Life-threatening airway obstruction is a dreaded complication of mediastinal masses. The acute management is difficult and catastrophic outcomes have been reported. We describe our experience in a Medical Intensive Care Unit (ICU) and discuss the management options.

Retrospective observational study.

1,600-bed tertiary-care hospital.

13 patients were treated between January 1996 and February 2010.

All patients were mechanically ventilated. Three received emergent chemotherapy, 10 underwent airway stenting and 2 underwent surgery.

The median age was 51 years (range 13 to 64). In 5 patients (36%), the diagnosis was only made after ICU admission. Two had benign pathologies (retrosternal goitre and bronchogenic cyst) and 3 had lymphoma. The remaining 8 had metastatic mediastinal lymphadenopathy. All were successfully weaned off the ventilator (range 0 to 5 days). All patients with benign pathologies and lymphoma were still alive with a mean follow-up of 2.8 years (range 1 to 7 years). All patients with metastatic disease died, with a mean survival of 3.8 months (range 1 to 9 months).

In the management of life-threatening airway obstruction by mediastinal masses, we reported the heterogeneous nature of this group of conditions. Therefore in the intensive care setting, a definitive diagnosis needs to be established urgently, as treatment and prognosis are highly dependent on the underlying etiology. We propose categorising patients into subgroups and combining various therapeutic modalities.

Point-of-care testing (POCT) for blood gas and chemistries is widely adopted in the emergency department (ED) for fast turnaround and critical medical decisions.

To assess performance of ED physicians compared to laboratory technologists carrying out i-STAT analysis. Impact of user skill on POCT performance was also studied.

Over a 3-month period, ED physicians performed tests with their i-STAT unit with parallel blood specimens (n=60) sent to the central laboratory to be tested on another i-STAT unit and laboratory analysers (blood gas and chemistries). Comparisons to laboratory results (obtained with ABL 520 and Beckman-Coulter LX20PRO) were carried out. Two operators from ED and central laboratory performed side by side, repeated tests on the i-STAT units. Their overall within-run imprecision were compared. The analytical performance of i-STAT was further assessed by another laboratory technologist to verify its capability to deliver good results.

Platform comparisons showed that ED physicians performed poorer compared to laboratory technologists, particularly with chloride and pO2 (Spearman coefficient of correlation — 0.49 and 0.54 respectively indicate wide scatter). This is also mirrored in the higher overall within-run imprecision for chemistries (sodium, chloride, potassium, glucose, urea and creatinine) by an ED physician (5.4%) compared to a laboratory technologist (3.7%). Including blood gas results showed an even wider gap in their testing skills. Except at a low creatinine of 67 μmol/L, the i-STAT analyser delivered precise measurements for the chemistries and blood gas parameters.

Performances by medical staff with the i-STAT analyser did not adequately approach the same skill level as laboratory technologists.

The aim of this pilot study was to assess the effects of music listening for pain relief among burns patients during the change of dressings.

Quasi-experimental study was conducted to evaluate if music helps in reducing the intensity of pain experienced by burns patients during wound dressing procedures. A convenience sample of 30 dressing changes were involved. Each dressing change was alternatively assigned to control (not listening to music) and treatment (listening to music) groups. Three assessment tools were used, namely: numeric pain scale, pain behavioural tool, and physiological monitoring. Patients who were given music to listen to were also asked about their experience at the end of the dressing change.

Most of the patients experienced little or no pain before dressing change, moderate pain during dressing change and lesser pain thereafter. There were no statistically significant differences in the pain experienced between wound dressing changed with no music listening and wound dressing changed with music listening. Despite a greater proportion of patients being distressed during wound dressing changes without music listening, there was no statistical evidence that music listening had an effect on the patient's distressed status. For wound dressing changes with music listening, patients experienced a lesser degree of change in mean blood pressure and heart rate (not statistically significant) as compared to wound dressing changes without music listening.

Preliminary results from this pilot study demonstrated that listening to music might have a positive effect on patient's experience of pain. Large studies are needed to evaluate the impact of music. However many patients verbalised that the soothing music helped to distract them from the pain experience during the dressing changes.

The study emergency department (ED) receives a new cohort of junior doctors every 6 months as part of the national manpower allocation to meet health service needs. An approach to effect a consistent reduction in prescribing errors by these doctors during their 6-month postings at the study ED was introduced in May 2009.

Monthly broadcasting of prescribing error counts by individual doctors allowed performances to be openly benchmarked against one another while providing a reference for self-improvement. The monthly tally of absolute counts were computed into a pre-determined formula and translated into monetary rewards for the junior doctors. Control charts of total count per month, mean count per doctor per month and highest individual count per month were plotted based on a total of 48 data points for 8 cohorts over 4 years from May 2007 to April 2011.

There was a shift towards fewer absolute counts and lower mean count of prescribing errors per month in the post-intervention phase, suggesting a positive impact of the intervention on the doctors as a cohort. The intervention's impact on individual performances, however, was less convincing with minimal change in variation in the highest individual count per month over the 4 years.

The use of metrics has helped to drive and sustain a reduction in prescribing errors among junior doctors as a cohort during their 6-month postings in the study ED.

To examine the major determinants influencing the decision to withdraw or continue life support and the attitudes of registered nurses towards the treatments for the critically ill patients in intensive care units (ICUs).

A self-administered questionnaire was distributed to a convenience sample of registered nurses working in 4 ICUs of a large teaching hospital in Singapore. The questionnaire consisted of 3 sections. The first section gathered details about the respondent's demographics; the second section required respondents to rate the importance of 18 different factors influencing the decision to withdraw or continue life support (1 = least important, 5 = most important) while the third section elicited responses about management strategy to 2 clinical scenarios from a possible pool of 6 different scenarios.

Eighty-three nurses responded (response rate: 70%). Patient advance directives (μ =3.63), likelihood of surviving current episode (μ = 3.52) and premorbid cognitive function (μ = 3.49) were the most important consideration factors for the withdrawal of life support. Across various clinical scenarios, the majority (82.6%) reported that “continue with current management” was most likely to be in the patients' best interests. Level of agreement between what the respondents believed to be in the patient's best interest and their responses on what they would do if they encountered a similar case was good (Kappa=0.78).

Respondents regarded “Patient advanced directives” to be the most important factor in influencing decisions to withdraw or continue life support. In general, nurses were uncomfortable with complete withdrawal of life support. One way to minimise the impact of clinician's social, ethical, moral and religious values on medical decision making is to encourage more Singaporeans to make an Advance Medical Directive.

Heparin has been used in flush solutions in attempts to prolong the patency of arterial and central venous pressure monitoring lines. Its use however is not without risks such as hypersensitivity reactions and heparin-induced thrombocytopenia. Recent studies suggest that non-heparinised solutions may be as effective in maintaining catheter patency. Our aim is to determine if there is any difference in maintaining patency of arterial and central venous pressure monitoring lines using normal saline as compared to heparinised saline.

In this randomised controlled trial, patients were randomised to receive either heparinised saline or normal saline flush solutions. Randomisation was achieved using sealed opaque envelopes. Only nurses who had been specially trained in the care of catheters cared for the lines. Data was collected at 8-hour intervals for up to 120 hours post-insertion on acceptability of arterial waveform and presence of backflow of blood in the arterial and central venous lines.

A total of 70 patients were enrolled, with 36 in the normal saline group and 34 in the heparinised saline group. There were no differences between the groups in functionality of the arterial catheters as determined by waveform and backflow at 72 hours (p=0.578) and 120 hours (p=0.328) as well as for the venous catheters at 72 hours (p=0.063) and at 120 hours (p=0.173) based on presence of backflow.

Normal saline is as effective as heparinised saline as flush solutions. Eliminating heparin use in flush solutions will result in fewer potential adverse events and cost savings.

Vasodilatory septic shock unresponsive to fluid resuscitation requires the addition of vasopressors. Catecholamines remain the first line vasopressor therapy, but treatment failure is a potential problem. Vasopressin and its analogue, terlipressin, have been used for this indication.

This systematic review aims to evaluate the effects of vasopressin and terlipressin on mortality and morbidity outcomes in patients with vasodilatory shock. Secondary outcomes include the effects of vasopressin and terlipressin on haemodynamic stability and organ function.

A computerised search of MEDLINE from January 1966 till June 2010 and screening of references of relevant articles were conducted. Only prospective, randomised controlled trials comparing vasopressin or terlipressin versus standard vasopressors or placebo were included.

Seven studies using vasopressin, three using terlipressin and one using both were identified. Four vasopressin trials assessing mortality and morbidity outcomes showed a trend towards benefit for mortality in adults but possibly adverse outcomes in a small paediatric study. No data was available on the long-term mortality and morbidity outcomes of terlipressin. Vasopressin and terlipressin were similar to standard vasopressors in maintaining haemodynamic parameters, while allowing a beneficial catecholamine-sparing effect. These agents also had a neutral to positive effect on organ function.

Vasopressin and terlipressin was comparable to conventional agents in the maintenance of haemodynamic stability and organ function in vasodilatory shock. Since morbidity and mortality data do not differentiate vasopressin and terlipressin from catecholamines, their role remains unclear. More large studies evaluating the long-term outcomes in this group of patients are required.