Variation in haemodynamic monitoring for major surgery in European nations: secondary analysis of the EuSOS dataset

More than 312 million surgical procedures are performed worldwide each year with an estimated mortality between 1 and 4 % [1, 2]. Postoperative complications and death are most frequent amongst high-risk patients, who are older, have co-morbid disease and undergo major surgery [3]. The dose of intravenous fluid and vasoactive drugs has an important effect on patient outcomes following major gastrointestinal surgery [4]. These treatments are prescribed according to subjective criteria leading to wide variation in clinical practice [5‐9]. One potential solution to this problem is the use of cardiac output monitoring to guide administration of intravenous fluid and vasoactive drugs [4].

The use of perioperative cardiac output monitoring remains controversial, with differing interpretations of the evidence base for this treatment approach [10‐18]. Recent evidence from a large clinical trial suggests that the benefit associated with this technology may be more marginal than previously believed, whilst some commentators have raised safety concerns [4, 19, 20]. Cardiac output monitoring has been recommended for patients undergoing selected types of major surgery both by the National Institute for Health and Care Excellence (NICE) in the UK and in a report commissioned by the Centers for Medicare and Medicaid Services in the USA [21, 22]. However, clinician surveys and anecdotal evidence suggest there is wide variation in the use of this technology [6‐9].

Across Europe, little is known about the use of cardiac output monitoring in non-cardiac surgery. A previously published 7-day cohort study described mortality and perioperative care in 28 European nations (European Surgical Outcomes Study [EuSOS]) [2]. Data were collected which describe the use of cardiac output monitoring and central venous catheterisation. We performed a secondary analysis of the EuSOS dataset to describe the prevalence and types of cardiac output monitoring in patients undergoing major surgery across European nations.

The principal finding of this analysis was that one in ten patients undergoing major surgery is exposed to cardiac output monitoring and one in five patients receives a CVC. Cardiac output monitoring and CVC use were more frequent with increasing ASA score and increasing urgency of surgery. The use of both cardiac output monitoring and CVCs was associated with a longer hospital stay, more admissions to critical care and higher mortality. Importantly, there was wide variation in the use of cardiac output monitoring and CVCs across European nations. The most commonly used method of cardiac output monitoring was arterial waveform analysis.

There is very little published patient-level data describing the use of cardiac output monitoring either in individual countries or at an international level. However, questionnaire-based surveys amongst anaesthesiologists have been used to explore the use of cardiac output monitoring in high-risk surgical patients. The most widely cited of these is an online questionnaire survey conducted in 2011 by Cannesson et al. amongst members of the American Society of Anesthesiologists and the European Society of Anaesthesiology. From 368 respondents, 35 % routinely monitored cardiac output with similar numbers in Europe and in North America [6]. A similar survey of members of the Korean Society of Anesthesiologists revealed that 59 % of 139 respondents were using cardiac output monitoring [8]. The findings of a paper-based survey of Chinese anaesthetists showed that cardiac output monitoring was used by 13 % of 210 respondents [9]. In addition to the variation in use of cardiac output monitoring, these surveys also reveal differences in the monitoring technique used. This variability is supported by another online questionnaire-based survey, exploring the use of goal-directed therapy in major elective surgery amongst members of the Association of Anaesthetists of Great Britain and Ireland, the American Society of Anesthesiologists and the Australia and New Zealand College of Anaesthetists [7]. Overall, the findings of these clinician surveys are consistent with the findings of our analysis of patient-level data. The lower use of cardiac output monitoring in our data compared to the data provided by Cannesson et al. might be due to the broader inclusion criteria of our study [6]. Suggested explanations for the wide variation in clinical practice were local factors including availability of equipment, experience or education and national factors like reimbursement and guidelines [6, 21‐23]. Involvement of clinicians in the development of, and therefore familiarity with, a certain device may also provide another explanation, which is why, for example, the use of PiCCO (PULSION Medical Systems, Germany) or LiDCO (LiDCO Ltd., United Kingdom) seems to be focused on their country of origin [6]. In addition, involvement of national opinion leaders in clinical trials improving the evidence base for certain devices might have contributed to the more extensive use of oesophageal Doppler monitoring in the United Kingdom compared to other nations [24‐26]. Scepticism regarding possible benefits and harms is seen as another important contributor [6]. This in view of studies which suggested that fluid restriction might be equally beneficial as fluid optimization [19, 27, 28].

To our knowledge, this is the largest available patient dataset describing the use of haemodynamic monitoring across international boundaries. However, the study only provides evidence of activity in one 7-day period in participating centres and so may not provide an accurate reflection of activity over a longer time frame. The data were collected 4 years ago, and contemporary practice may have evolved since the original study. The small number of patients returned by a small number of hospitals in smaller countries may also have resulted in bias and a poor representation of patterns of monitoring use within those countries. In addition, the small patient numbers in some sub-categories limit the robustness of any analysis of variation between hospitals or potential clustering within hospitals, especially as some centres may specialise in particular types of surgery with consequent effects on care pathways and case mix. The study was not specifically designed for collection of data describing the use of haemodynamic monitoring and did not provide precise detail on the specific product used in each patient. No data were collected to describe how haemodynamic data were used, the target values for specific variables or which interventions were used to attain these targets. In addition, there is no data describing the indication for CVC placement. Due to the heterogeneity of cases and the limited sample size, we were not able to further analyse effects of haemodynamic monitoring on clinical outcomes.

One in ten patients undergoing major surgery in Europe is exposed to cardiac output monitoring whilst one in five receives a central venous catheter. The use of both technologies varies widely across nations. Further research is needed to confirm the clinical value of cardiac output monitoring and how this technology might help to further reduce postoperative morbidity and mortality.