1 Anaesthesiology Division and 2 Quality of Care Unit, Geneva University Hospitals, Rue Micheli-du-Crest 24, CH-1211 Geneva 14, Switzerland*Corresponding author
Accepted for publication: May 30, 2002
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Method. The investigation was based on a framework specifically developed for the analysis of medical accidents. This framework helped to identify the chronology and outcome of the case, the care management problems and the factors that led to the event. Information was obtained by interviewing the anaesthesiologist in charge of the patient.
Results. Occlusion was not recognized because the filter was hidden under the drapes and below the patients head. To reduce the frequency of this event, we recommend that filters should be visible, placed above the level of the patients body, or mounted on the expiratory circuit, at a distance from patients airway.
Conclusions. To allow appropriate corrective actions, critical incidents and accidents should be systematically investigated using root-cause analysis.
Br J Anaesth 2002; 89: 6335
Keywords: audit, root cause analysis; ventilation, respiratory circuit; risk
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
To illustrate the potential of such an approach, we describe the root-cause analysis of a case of filter occlusion by water and secretions during anaesthesia that could have been fatal.
![]() |
Methods and results |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
To identify the root cause of the event, we used a framework developed for the investigation of medical accidents.2 We first established the chronology and the outcome of the case. We then identified care management problems, which were defined as either equipment failures or active failures represented by deviations beyond safe limits. Finally, we listed and categorized the different factors that facilitated the occurrence of these problems. A contributory factor was considered specific when it was related to chance and general when it was a permanent feature of the system and as such represented a root cause of the incident. Information was obtained by interviewing the anaesthesiologist in charge of the case.
To devise corrective actions, we relied on two principles that are commonly combined when managing risks, namely prevention and absorption.3 Prevention aims at removing the root causes which increase the opportunity of equipment or active failures whereas absorption is intended at eliminating root causes that hinder the early identification and correction of active failures.
Chronology and outcome
A 13-yr-old boy with maxillo-facial Cacrum Oris (Noma) was to have elective plastic reconstruction. Patient history, examination, and laboratory findings were otherwise normal.
During surgery, the respiratory system consisted of a ventilator connected via a breathing tubing to an airway pressure and gas analyser and a Y-piece, fitted to a BB 25 airway filter (Pall Medical, Ann Arbor, MI, USA). The filter was connected to a catheter mount tubing connector and a tracheal tube. The ventilator was checked before surgery. To provide convenient access to the surgical site, the filter was positioned below the patients head under the drapes.
The beginning of the procedure was uneventful but after 2 h, the peak airway pressure progressively increased, exceeding the 30 cm H2O threshold. Both ventilator and analyser pressure alarms sounded. The junior anaesthesiologist in charge of the case immediately increased the inspired oxygen concentration. Noisy chest sounds were heard on auscultation, which decreased after large amounts of secretion were removed using a suction catheter inserted into the tracheal tube. However, the peak airway pressure remained greater than 40 cm H2O. Four minutes after the alarms were triggered, inflation of the patients lungs became impossible. Oxygen saturation decreased to 80%. When the junior anaesthesiologist connected a self-inflating bag directly to the tracheal tube, immediate lung ventilation was possible. The ventilator and the respiratory circuit were then replaced and the tracheal tube was reconnected using the same analyser, filter, and tubing connector. Ventilation remained impossible. Finally, occlusion of the filter by water and secretions was identified and the filter was changed. The patient recovered quickly and without sequel.
Identification of care management problems and contributory factors
In this critical incident, two care management problems occurred, namely an equipment failure (the occlusion of the airway filter), and its delayed identification. These two events, the combination of which led to severe hypoxaemia, were facilitated by a set of specific or general contributory factors presented in Table 1.
|
To prevent liquids from accumulating in the filter, it was decided that the filter should be routinely placed above the lung level or, when this was not possible it should be mounted on the expiratory circuit at a distance from patients airways.
New instructions for use of filters were printed on posters and displayed in all operating rooms.
![]() |
Comment |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
In most cases, the problem is not the individual but the system. Our analysis shows that occlusion was possible because the filter had a hydrophobic membrane and was placed below the patients head. The ability of the anaesthesiologists to detect airway filter occlusion was limited by the lack of information on the risk associated with the use of this device and by its position under the drapes. The identification of latent failures that are organizational root causes of accidents or critical incidents is, therefore, essential for devising appropriate corrective actions.
At first glance, our strategy, which combines prevention and absorption, could be perceived as convoluted. Compared with the prevailing approach of risk management which, as noted by Cooper,6 depends almost solely on the anaesthetists ability to react instinctively and flawlessly every time a problem arises, this method is much more efficient. Indeed, if Po is the probability of airway filter occlusion and Pr the probability of not recognizing in time the problem, PoxPr will be the probability Pc of the critical incident. As a consequence, any x-fold decrease in both Po and Pr will have a multiplicative effect and result in a x2-fold decrease of Pc whereas a x-fold decrease in Pr alone will only cause an equivalent decrease in Pc.
Given the usefulness of root-cause analysis with regard to patient safety, it is clear that we should systematically investigate critical incidents and accidents. However, reporting and analysis of such events on a routine basis and setting up appropriate corrective actions will require considerable cultural and organizational changes. Is there any alternative?
![]() |
Acknowledgement |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Vincent CA, Taylor-Adams S, Chapman EJ, et al. A Protocol for the Investigation and Analysis of Clinical Incidents. London: University College London/Association of Litigation and Risk Management, 1999
3 Leape LL. Error in medicine. JAMA 1994; 272: 18517[ISI][Medline]
4 Norman PH, Daley MD, Walker JR, Fusetti S. Obstruction due to retained carbon dioxide absorber canister wrapping. Anesth Analg 1996; 83: 4256[ISI][Medline]
5 Aarhus D, Soreide E, Holst-Larsen H. Mechanical obstruction in the anaesthesia delivery-system mimicking severe bronchospasm. Anaesthesia 1997; 52: 9924[ISI][Medline]
6 Cooper JB, Newbower RS, Kitz RJ. An analysis of major errors and equipment failures in anesthesia management: considerations for prevention and detection. Anesthesiology 1984; 60: 3442[ISI][Medline]