Abstract
Most procedures in cardiac catheterisation laboratories (CCLs) have traditionally been performed under conscious sedation under the supervision of the treating proceduralist. With growing demand for more complex procedures to be performed, in emergencies and in patients with limited cardiorespiratory reserve, a reconsideration of the level of supervision provided is required. We conducted a retrospective cohort study of all patients who had CCL procedures and required an overnight stay at Royal North Shore Hospital during a 12-month period prior to introducing monitored anaesthesia care (MAC), compared with a 12-month period following introduction of MAC on selected weekdays. Primary outcomes were the difference in rates of Code Blue calls (triggering the cardiac arrest team), Clinical Reviews and Rapid Responses (defined as per the NSW Health ‘Between the Flags’ Deteriorating Patient Safety Net System) in the 24 h post-procedure between patients who did and did not have MAC. The secondary outcome was a difference in mortality (within 24 h of a procedure and in-hospital) between patients who did and did not have MAC. One thousand nine hundred and eight patients were analysed (926 pre-intervention, 982 post-intervention). We found no statistically significant difference in any of the primary or secondary outcomes between the pre-intervention and post-intervention patients overall. However, we found a statistically significant lower rate of Code Blue calls in patients who had MAC (n = 3, 0.6%) compared with no MAC (n = 31, 2.3%). We also found a significantly lower 24-h mortality in patients who had MAC (n = 1, 0.2%) compared with no MAC (n = 22, 1.6%), but no difference in overall in-hospital mortality.
Keywords
Introduction
In Australia, the cardiac catheterisation laboratory (CCL) is the site for most minimally invasive cardiac procedures. Historically, sedation for these procedures occurred under the direction of the treating cardiologist, without the routine presence of an anaesthetist. With clinical advances seen in the last 30 years, the procedures being undertaken in the CCL are increasingly diverse and complex, and are being performed on elderly, comorbid patients with limited cardiorespiratory reserve. 1 Monitored anaesthesia care (MAC) has been introduced in many institutions in response to this higher-risk cohort. However, MAC is yet to be standard for all patients having CCL procedures in Australia.
MAC refers to the presence of a qualified anaesthetist throughout a procedure, who oversees continuous cardiorespiratory monitoring, tailored drug selection, and is capable of early identification and escalation in cases of patient deterioration. 1 The alternative to MAC is commonly referred to as nurse-led sedation (NLS). Evidence to support MAC in the CCL is limited, and most published literature is restricted to examining the safety of different sedation models for patients undergoing transcatheter aortic valve implantation (TAVI).2 –4 Three single-centre non-randomised studies conducted overseas failed to show improved safety outcomes for MAC compared with NLS in the TAVI population.2 –4 However, there are methodological limitations to these studies, including small sample sizes and assigning high-risk patients to the MAC group.2,4 In addition, TAVI implies a complicated procedure in a high-risk patient, and these results cannot be generalised to the whole CCL population. Furthermore, TAVI cases are already largely accepted as procedures warranting MAC. There is an urgent need for studies examining the impact of MAC in the CCL generally, including for lower-risk procedures such as diagnostic angiography.
There are two published guidelines related to the provision of sedation in the CCL in Australia. The Australian and New Zealand College of Anaesthetists Guideline on Procedural Sedation recommends the presence of at least the proceduralist, an anaesthetist (or doctor trained in requisite sedation competencies) and a third assistant who may be a registered nurse. 5 The position statement from the Cardiac Society of Australia and New Zealand recommends the presence of two staff members: the proceduralist and an assistant (sedationist, junior medical officer or registered nurse) to administer the sedation, monitor the patient and provide procedural assistance. 6
Our institution is a large tertiary hospital in Sydney, Australia, that currently performs nearly 1000 inpatient CCL procedures per year. At our centre, the CCL is located approximately 100 metres away from the theatre complex, and until recently was not routinely staffed by anaesthetists. Cases deemed high risk, such as structural heart procedures, could have oversight by an anaesthetist upon request. In response to a series of adverse events, our hospital initiated a trial of routine MAC in the CCL on selected weekdays. We performed a retrospective cohort study to assess the impact of introducing MAC on the rates of Code Blue calls, Clinical Reviews and Rapid Responses and all-cause in-hospital patient mortality (within 24 h and in-hospital).
Methods
Study population
All patients who had a procedure in the CCL at our hospital followed by an inpatient admission for at least one night post-procedure were included in the study. Standard practice at our institution is for patients undergoing elective, uncomplicated coronary angiography to be discharged home the same day (day-case patients). These patients were excluded from the study given the absence of outcome data. Patients were also excluded if their disposition post-procedure was to another hospital, most commonly the private hospital associated with our facility, given the heterogeneity in postoperative monitoring and difficulty comparing outcome data.
All CCL procedure types, including angiography, electrophysiology studies, structural and valvular procedures, among others, were included in the study. The standard procedure at our hospital was for same-day discharge home for patients having elective angiography. Therefore, many of these low-risk patients were excluded from the study. However, if there was any periprocedural concern they would be admitted and included in our study.
Intervention
Monitored anaesthesia care was introduced to the CCL in April 2021. This included the presence of a consultant anaesthetist and an accredited dedicated anaesthetic nurse. Equipment available included all airway, continuous haemodynamic monitoring and drug supplies expected to be on standardised anaesthetic trolleys in Australian tertiary hospitals. The drugs, level of sedation and airway management were decided by the treating anaesthetist with consideration given to the patient and the procedure.
Comparison
In the absence of MAC, a cardiology nursing staff member delivered sedation as directed by the cardiologist proceduralist, which would normally be boluses of fentanyl and midazolam. For high-risk cases performed prior to April 2021, an anaesthetist could be specifically arranged. This was usually only organised for structural heart cases on an as-needed basis, with no regular anaesthetic support outside of this. Given the very small numbers of structural cases performed each year, the whole of the pre-intervention group was considered to have had their procedure without MAC.
Groups
The pre-intervention group was all patients meeting the inclusion criteria who had their CCL procedure in the 12-month period between 1 February 2020 and 31 January 2021.
In April of 2021, MAC was introduced in the CCL on Tuesday, Wednesday and Thursday. The decision to limit routine MAC to three days was based on funding and staff availability, along with perceived demand based on the number of adverse clinical events occurring in the CCL.
The post-intervention group was all patients meeting the inclusion criteria who had their CCL procedure in the 12-month period between 1 June 2021 and 31 May 2022. This was to minimise any effect from the change associated with the initial rollout of providing MAC. During this post-intervention period, patients who had their procedure performed on a Tuesday, Wednesday or Thursday (‘MAC days’) received MAC regardless of their acuity or risk. Procedures falling on other days of the week (‘non-MAC days’) continued without routine MAC.
Outcomes
The primary outcome was to determine the difference in the rate of Code Blue calls, Clinical Reviews and Rapid Responses in the 24 h following a CCL procedure between patients who did and did not have MAC. The secondary outcome was to determine the difference in mortality rate within 24 h of procedure, and mortality at any time during the inpatient admission (in-hospital mortality) between patients who did and did not have MAC during their procedure.
A Code Blue call was defined as an arrest call which is initiated by pressing an emergency button. At our hospital, Code Blue calls are reserved for critically unstable peri-arrest or arrested patients, and they activate a specialised team with emergency equipment such as airway devices and pacing capabilities. While ‘false’ Code Blue calls are infrequent, our data included a description of the reason for the code, and any ‘false’ calls where the button had been pressed accidently were excluded.
Clinical Reviews and Rapid Responses are based on the NSW Health ‘Between the Flags’ Deteriorating Patient Safety Net System, based on vital sign observations and/or additional criteria. 7 The extent of haemodynamic instability dictates the type of review called by nursing staff. Mildly deranged haemodynamic values trigger a Clinical Review, to be attended by a junior medical officer from the treating team (during hours) within 30 min. More unstable haemodynamic values trigger a Rapid Response, attended by a junior medical officer and a registrar from the intensive care unit within 10 min. The Between the Flags system is an important method for early identification of deterioration, allows for escalation of care and provides objective and reproducible outcome measures.
Data collection
Patients who had a CCL procedure within the included date ranges were identified from the electronic CCL database. Data elements available were procedure date, procedure type, discharge destination and de-identified patient medical record number. The study population was then linked to a hospital database of all Code Blue calls, Clinical Reviews, Rapid Responses and in-hospital mortality. We did not perform chart reviews so do not have access to demographic data or specific information regarding the anaesthetic, cause of deterioration or death. The study was approved by the Northern Sydney Local Health District Human Research Ethics Committee, 2022/ETH02612.
Statistical analysis
Between-group differences were analysed using chi-square test (Fisher’s exact test when the cell number was less than 5) for categorical differences. Results were considered significant if the P-value was less than 0.05. Statistical analyses were performed using an online statistical package. 8 No a priori sample size calculation was performed as this was a historical cohort study.
Results
Study population
A total of 1908 patients were included (926 in the pre-intervention group and 982 in the post-intervention group). This reflects approximately one-quarter of total CCL cases performed in each 12-month period (3610 in the pre-intervention period and 3270 in the post-intervention period). Table 1 presents the procedure types for patients who received MAC, compared with no MAC, noting that several patients had two procedure types performed during a single CCL visit. The most common procedure type in both groups was coronary angiography with or without percutaneous intervention.
Cardiac catheterisation laboratory procedure types.
Heart study = structural heart procedure.
Includes balloon aortic valvuloplasty, Pascal/Mitraclip mitral valve repair, TAVI, valve-in-valve replacement.
MAC: monitored anaesthesia care; PCI: percutaneous coronary intervention; TAVI: transcatheter aortic valve implantation.
Table 2 shows the primary and secondary outcomes for the pre-intervention and post-intervention groups. When comparing the pre-intervention group with the post-intervention group (including all days of the week, MAC days and non-MAC days) we did not find any significant difference in any outcome measures.
Primary and secondary outcomes comparing the pre-intervention with the post-intervention groups.
CCL: cardiac catheterisation laboratory.
Table 3 shows the rates of primary and secondary outcomes post-intervention for the MAC days group and the non-MAC days group. When comparing patients who had MAC with those who did not receive MAC, we found a statistically significantly reduced rate of Code Blue calls and 24-h post-procedure mortality. The difference was significant when the MAC days group was compared with both the non-MAC days group and the total group who did not receive MAC (pre-intervention group and non-MAC days group). We failed to show significant differences for the other measured outcomes.
Primary and secondary outcomes comparing the MAC days group with the non-MAC days group and the total no MAC group.
MAC: monitored anaesthesia care
P < 0.05 compared with MAC days group.
Table 4 shows the sub-group analysis results based on procedure type, comparing diagnostic angiography (with or without percutaneous intervention) with all other procedure types. We found that the significant difference in rates of Code Blue calls and 24-h post-procedure mortality was also found in the angiography patient cohort.
Primary and secondary outcomes comparing MAC days with non-MAC days and total no MAC groups with sub-group analysis by procedure type.
Excluding cases that had another procedure during the same anaesthetic as for their angiography.
*P < 0.05 compared with MAC-days group.
MAC, monitored anaesthesia care; PCI: percutaneous coronary intervention.
Discussion
We performed a retrospective cohort study of inpatients undergoing a procedure in the CCL before and after the introduction of routine MAC three days a week. Monitored anaesthesia care indicates the provision of a specialist anaesthetist dedicated to providing tailored critical care management to unwell patients. This includes identifying deteriorating patients and escalating their care. It is unsurprising that MAC was associated with a significant improvement in rates of Code Blue calls and 24-h post-procedure mortality at our institution (Tables 3 and 4). This difference was also seen in the sub-group analysis of angiography patients with or without PCI. We did not find a significant association between MAC and any of the other primary or secondary outcomes.
When comparing the pre-intervention group with the whole of the post-intervention group, there was no significant difference in rates of primary or secondary outcomes (Table 2). A significant difference becomes apparent when comparing patients who received MAC (MAC days group within the post-intervention group) with those who did not receive MAC. This finding relates to comparing the MAC days group with both the non-MAC days post-intervention group and the total no MAC group (pre- and post-intervention composite), as shown in Table 3. This suggests that the benefit in patient outcomes is MAC-related rather than a product of improved institutional experience over time.
Within the post-intervention period, rates of 24-h post-procedure mortality were 2.1% in the non-MAC days group compared with 0.2% in the MAC days group, a significant difference (Table 3). In the absence of formal chart review we are unable to comment on the causes of death, nor, in particular, sedation-related factors associated with mortality. The significant difference in rates of both Code Blue calls and 24-h post-procedure mortality was also apparent in the sub-group analysis of patients undergoing angiography, but not in the other procedure types. The other procedures included valvular and structural heart procedures which are generally higher risk, and many occurred on Tuesday, Wednesday and Thursday. As such, it is not surprising that the difference in rates of study outcomes within this sub-group has been moderated by the scheduling of higher-risk cases to MAC days. This finding suggests that MAC may provide an important patient safety advance for all CCL patients and procedures.
Without performing chart review, we are assuming that the acuity of angiography patients in the MAC days group and non-MAC days groups was similar, given that the distribution of case numbers across days remained even. However, if higher-risk cases were deliberately rescheduled to fall on a MAC day in the post-intervention period, it is possible that this was a higher-risk cohort. If so, this only strengthens the suggestion that the presence of an anaesthetist improves patient outcomes, assuming that the improved rates of Code Blue calls and 24-h mortality for the MAC days group occurred despite a higher-risk population. Further studies incorporating risk data such as the American Society of Anesthesiologists Physical Status Classification are required to confirm this assumption.
Our results found a significant improvement in 24-h post-procedure mortality in the MAC days group, but, importantly, there was no significant difference in overall in-hospital mortality between the MAC days group and non-MAC days group (Table 3). We suggest that this is a product of the nature of MAC; anaesthetists identify, manage and escalate the care of deteriorating patients in the periprocedural period, but have little to do with the patients beyond 24 h postoperatively. For example, it is realistic that unwell patients who had MAC were more likely to be identified and referred to the intensive care unit than an equally unwell patient who received NLS, thus impacting the 24-h mortality rate more than the in-hospital mortality. It is also possible that higher acuity care in the first 24 h only delayed what may have been inevitable deaths related to patients’ underlying conditions.
The findings discussed above differ from previous literature that failed to find an improvement in patient outcomes with MAC for TAVI procedures.2 –4 A possible explanation was the use of risk stratification, with more complex patients being assigned to the MAC group, resulting in significant differences in groups’ characteristics despite propensity score matching.2,3 The largest of the studies reported a difference in in-hospital mortality (P = 0.0496) but failed to list this as a key finding despite predefining a 0.05 level of significance. 2
Unlike the TAVI studies, we also examined other procedures including angiography, and found that the significant association also applied to the angiography cohort, a generally lower-risk cohort not traditionally deemed in need of MAC. Also, our findings pertain only to patients requiring an in-hospital stay following their angiography, as we did not include the day cases which made up the majority of CCL procedures (more than 70%). Our case numbers were too small to be adequately powered to perform sub-group analyses for other procedure types such as valvular procedures. However, the trend is towards a reduced rate of all adverse outcomes measured except Clinical Reviews following introduction of MAC. This is an important finding given that angiography constitutes a vast majority of inpatient CCL cases at our institution, and providing MAC for this procedure type would be a significant patient safety update compared with current standard of care.
Our study has several limitations to be acknowledged. First, we relied on routinely collected health data and, as such, we must accept that the data may be incomplete or have errors from the data entry process. We also investigated only CCL patients requiring an overnight stay. In the absence of detailed chart review we are unable to comment on any relationships between demographic factors, choice of anaesthetic agents or cause of death. Furthermore, the retrospective nature of the study means we are unable to infer causation between the provision of MAC and any improved patient outcomes. Finally, as a single-centre study, the generalisability of our results to other centres needs to be confirmed.
Conclusion
We performed a retrospective cohort study of patients undergoing procedures in the CCL requiring an overnight stay and found that the provision of MAC was associated with a lower rate of Code Blue calls and 24-h post-procedure mortality. In contrast we found no statistically significant differences in the rates of Clinical Reviews, Rapid Responses or overall in-hospital mortality. Our results support the reasonable assumption that MAC would improve patient safety and outcomes, but are especially significant for demonstrating lower rates of Code Blue calls and 24-h post-procedure mortality (but not in-hospital mortality) within an angiography population who required an overnight stay. Further research is required to confirm our findings and elucidate the relationship with other variables, including patient risk factors, proceduralist skillset, sedation methods and causes of death or deterioration with or without MAC. If MAC was to be mandated in the CCL, as it has been for other sedation-based procedures such as endoscopy and colonoscopy, it will undoubtedly place additional demands on an already stretched healthcare system.
