doi:10.1111/codi.12487
Original article
Decision-making in rectal surgery E. MacDermid*, C. J. Young†, J. Young‡ and M. Solomon† *Department of Surgery, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia, †Department of Colorectal Surgery, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia and ‡Surgical Outcomes and Research Centre, Sydney University, Camperdown, New South Wales, Australia Received 21 June 2013; accepted 11 September 2013; Accepted Article online 27 November 2013
Abstract Aim The decision to create a stoma after anterior resection has significant consequences. Decisions under uncertainty are made with a variety of cognitive tools, or heuristics. Past experience has been shown to be a powerful heuristic in other domains. Our aim was to identify whether the misfortune of recent anastomotic leakage or surgeon propensity to take everyday risks would affect their decision to defunction a range of anastomoses. Method Questionnaires were sent to members of the Colorectal Surgical Society of Australia and New Zealand. Participants were asked for demographic information, questions regarding risk-taking propensity, when their last anastomotic leakage occurred and whether they would defunction a range of hypothetical rectal anastomoses grouped according to height, American Society of Anesthesiologists grade and use of preoperative radiotherapy. Scores were derived for hypothetical patient likelihood of having a stoma created and individual surgeon propensity for stoma formation. Hazard regression analysis was used to assess demographic predictors of stoma formation.
Introduction A familiar dilemma facing surgeons who perform anterior resection is whether or not to create a defunctioning stoma. Meta-analysis has shown that construction of a loop ileostomy significantly reduces the risk of symptomatic anastomotic leakage in low anterior resection [1]. Anastomotic leakage has been demonstrated to be the biggest single risk factor for death following anterior resection and has also been associated with a long-term reduction in several domains of patients’ quality of life, in particular physical functioning [2], pain and fatigue [3]. Correspondence to: Dr Ewan MacDermid, Department of Surgery, Royal Prince Alfred Hospital, Camperdown, New South Wales 2050, Australia. E-mail: [email protected]
Results In total, 110 (75.3%) of 146 surveyed surgeons replied; 72 (65.5%) reported anastomotic leakage within the last 12 months. Surgeons’ propensity for risk-taking was comparable (24.6 vs 27.53, 95% confidence interval, Mann–Whitney-U) to previously studied participants in economic models. Surgeon age (< 50 years) and lower propensity for risk-taking were demonstrated to be independent predictors of stoma formation on regression analysis. Conclusion Although the decision to create a stoma after anterior resection may be made in the belief that its foundation derives from rational thought, it appears that other unrecognized operator factors such as age and risk-taking exert an effect. Keywords Heuristics, stoma, rectal surgery What does this paper add to the literature? There is a relative lack of literature examining heuristics and decision-making in surgery and most of what is available is referenced in this paper. To the best of our knowledge there are no studies examining operator factors in stoma formation, thus making this paper both unique and relevant.
Formation of a defunctioning stoma, however, is not appropriate for every patient undergoing anterior resection. Temporary stoma formation invariably results in delayed discharge from hospital and has been associated with a complication rate of up to 42% [4], notably water and electrolyte derangement. It may also impact adversely on patients’ quality of life upon discharge [5,6]. When making the decision whether or not to defunction a patient, the surgeon must therefore balance the risk of anastomotic leakage and its consequences with the risk of complication and inconvenience of a stoma. This decision and its outcome can be summarized as in Fig. 1, a decision tree of the various end-points. The tools that surgeons use to arrive at such a decision include rational and irrational factors, as recognized
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 16, 203–208
203
Decision-making in rectal surgery
No stoma
Anterior resection
Defunctioning stoma
E. MacDermid et al.
No leak uncomplicated recovery
Reoperation
Leak
Non-operative management
No leak, no complications
Death
Complications of stoma +/– delayed closure
Reoperation
Leak
Non-operative management
Death
by Detmer et al. [6] and Slovic et al. [7]. While the surgeon may be influenced by rational factors in his or her decision-making, irrational factors may also come into play. The aims of the study were to identify patient and surgeon factors which would influence colorectal surgeons’ decision to defunction a range of hypothetical rectal anastomoses. Surgeon factors of particular interest were the misfortune of a recent anastomotic leak and previous risk-taking behaviour.
Method A self-administered questionnaire for surgeons was developed and posted to all registered members of the Colorectal Society of Australia and New Zealand. Participants were asked for demographic information including their age, gender, number of rectal anastomoses performed every year and when their last anastomotic leak occurred. They were also asked a series of questions derived from a previously published, validated questionnaire on risk-taking across several domains [8]. These included personal health and safety, gambling habits and social risks such as standing for office or leaving employment without an alternative. Participants’ propensity for risk-taking was then compared with that of previously studied professionals in business and finance [8]. The second part of the questionnaire asked the participants whether they would construct a defunctioning stoma for a variety of 30 clinical scenarios involving a hypothetical rectal anastomosis, grouped according to anastomotic height, patient American Society of Anesthesiologists (ASA) grade, age and whether or not they had received preoperative radiotherapy.
204
Figure 1 Decision tree of the various outcomes.
Ethical approval for the study was obtained from Sydney South West Area Health Service. Participants were deemed eligible to be in the study if they were members of the Colorectal Surgical Society of Australia and New Zealand and were currently in clinical practice. All potential participants were allocated a unique identifier number to enable monitoring of response rate. Questionnaires were de-identified prior to analysis. Statistical analysis
Data were analysed using MEDCALCTM statistical software (MedCalc Software, Ostend, Belgium). Each individual surgeon respondent was given a score derived from the total number of stomas they created. Similarly, each hypothetical patient scenario had a score derived from the number of surgeon respondents choosing to defunction it. Multiple regression analysis was used to assess the effect of individual prognostic patient and surgeon factors on the likelihood of stoma formation. A P-value of 0.05 or less was taken as the level of significance.
Results One hundred and ten (75.3%) of 146 surveyed surgeons responded to the questionnaire. All respondents reported having had an anastomotic leak at some time in their career, with 35.5% reporting one in the last 3–12 months (Table 1). Surgeons’ measured propensity for risk-taking was statistically comparable (24.6 vs 27.53, 95% confidence interval, Mann–Whitney-U) to previously studied participants in business and finance research [8] (Table 2). Across the 30 hypothetical rectal anastomoses scenarios the percentage of respondents choosing to defunction
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 16, 203–208
Decision-making in rectal surgery
E. MacDermid et al.
Table 1 Demographics of responders. n (%) Gender Male 103 (94) Female 7 (6) Age (years) 30–39 16 (15.7) 40–49 56 (50.9) 50–59 25 (22.7) 60+ 13 (11.8) Number of rectal anastomoses performed per annum
Original article
Decision-making in rectal surgery E. MacDermid*, C. J. Young†, J. Young‡ and M. Solomon† *Department of Surgery, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia, †Department of Colorectal Surgery, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia and ‡Surgical Outcomes and Research Centre, Sydney University, Camperdown, New South Wales, Australia Received 21 June 2013; accepted 11 September 2013; Accepted Article online 27 November 2013
Abstract Aim The decision to create a stoma after anterior resection has significant consequences. Decisions under uncertainty are made with a variety of cognitive tools, or heuristics. Past experience has been shown to be a powerful heuristic in other domains. Our aim was to identify whether the misfortune of recent anastomotic leakage or surgeon propensity to take everyday risks would affect their decision to defunction a range of anastomoses. Method Questionnaires were sent to members of the Colorectal Surgical Society of Australia and New Zealand. Participants were asked for demographic information, questions regarding risk-taking propensity, when their last anastomotic leakage occurred and whether they would defunction a range of hypothetical rectal anastomoses grouped according to height, American Society of Anesthesiologists grade and use of preoperative radiotherapy. Scores were derived for hypothetical patient likelihood of having a stoma created and individual surgeon propensity for stoma formation. Hazard regression analysis was used to assess demographic predictors of stoma formation.
Introduction A familiar dilemma facing surgeons who perform anterior resection is whether or not to create a defunctioning stoma. Meta-analysis has shown that construction of a loop ileostomy significantly reduces the risk of symptomatic anastomotic leakage in low anterior resection [1]. Anastomotic leakage has been demonstrated to be the biggest single risk factor for death following anterior resection and has also been associated with a long-term reduction in several domains of patients’ quality of life, in particular physical functioning [2], pain and fatigue [3]. Correspondence to: Dr Ewan MacDermid, Department of Surgery, Royal Prince Alfred Hospital, Camperdown, New South Wales 2050, Australia. E-mail: [email protected]
Results In total, 110 (75.3%) of 146 surveyed surgeons replied; 72 (65.5%) reported anastomotic leakage within the last 12 months. Surgeons’ propensity for risk-taking was comparable (24.6 vs 27.53, 95% confidence interval, Mann–Whitney-U) to previously studied participants in economic models. Surgeon age (< 50 years) and lower propensity for risk-taking were demonstrated to be independent predictors of stoma formation on regression analysis. Conclusion Although the decision to create a stoma after anterior resection may be made in the belief that its foundation derives from rational thought, it appears that other unrecognized operator factors such as age and risk-taking exert an effect. Keywords Heuristics, stoma, rectal surgery What does this paper add to the literature? There is a relative lack of literature examining heuristics and decision-making in surgery and most of what is available is referenced in this paper. To the best of our knowledge there are no studies examining operator factors in stoma formation, thus making this paper both unique and relevant.
Formation of a defunctioning stoma, however, is not appropriate for every patient undergoing anterior resection. Temporary stoma formation invariably results in delayed discharge from hospital and has been associated with a complication rate of up to 42% [4], notably water and electrolyte derangement. It may also impact adversely on patients’ quality of life upon discharge [5,6]. When making the decision whether or not to defunction a patient, the surgeon must therefore balance the risk of anastomotic leakage and its consequences with the risk of complication and inconvenience of a stoma. This decision and its outcome can be summarized as in Fig. 1, a decision tree of the various end-points. The tools that surgeons use to arrive at such a decision include rational and irrational factors, as recognized
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 16, 203–208
203
Decision-making in rectal surgery
No stoma
Anterior resection
Defunctioning stoma
E. MacDermid et al.
No leak uncomplicated recovery
Reoperation
Leak
Non-operative management
No leak, no complications
Death
Complications of stoma +/– delayed closure
Reoperation
Leak
Non-operative management
Death
by Detmer et al. [6] and Slovic et al. [7]. While the surgeon may be influenced by rational factors in his or her decision-making, irrational factors may also come into play. The aims of the study were to identify patient and surgeon factors which would influence colorectal surgeons’ decision to defunction a range of hypothetical rectal anastomoses. Surgeon factors of particular interest were the misfortune of a recent anastomotic leak and previous risk-taking behaviour.
Method A self-administered questionnaire for surgeons was developed and posted to all registered members of the Colorectal Society of Australia and New Zealand. Participants were asked for demographic information including their age, gender, number of rectal anastomoses performed every year and when their last anastomotic leak occurred. They were also asked a series of questions derived from a previously published, validated questionnaire on risk-taking across several domains [8]. These included personal health and safety, gambling habits and social risks such as standing for office or leaving employment without an alternative. Participants’ propensity for risk-taking was then compared with that of previously studied professionals in business and finance [8]. The second part of the questionnaire asked the participants whether they would construct a defunctioning stoma for a variety of 30 clinical scenarios involving a hypothetical rectal anastomosis, grouped according to anastomotic height, patient American Society of Anesthesiologists (ASA) grade, age and whether or not they had received preoperative radiotherapy.
204
Figure 1 Decision tree of the various outcomes.
Ethical approval for the study was obtained from Sydney South West Area Health Service. Participants were deemed eligible to be in the study if they were members of the Colorectal Surgical Society of Australia and New Zealand and were currently in clinical practice. All potential participants were allocated a unique identifier number to enable monitoring of response rate. Questionnaires were de-identified prior to analysis. Statistical analysis
Data were analysed using MEDCALCTM statistical software (MedCalc Software, Ostend, Belgium). Each individual surgeon respondent was given a score derived from the total number of stomas they created. Similarly, each hypothetical patient scenario had a score derived from the number of surgeon respondents choosing to defunction it. Multiple regression analysis was used to assess the effect of individual prognostic patient and surgeon factors on the likelihood of stoma formation. A P-value of 0.05 or less was taken as the level of significance.
Results One hundred and ten (75.3%) of 146 surveyed surgeons responded to the questionnaire. All respondents reported having had an anastomotic leak at some time in their career, with 35.5% reporting one in the last 3–12 months (Table 1). Surgeons’ measured propensity for risk-taking was statistically comparable (24.6 vs 27.53, 95% confidence interval, Mann–Whitney-U) to previously studied participants in business and finance research [8] (Table 2). Across the 30 hypothetical rectal anastomoses scenarios the percentage of respondents choosing to defunction
Colorectal Disease ª 2013 The Association of Coloproctology of Great Britain and Ireland. 16, 203–208
Decision-making in rectal surgery
E. MacDermid et al.
Table 1 Demographics of responders. n (%) Gender Male 103 (94) Female 7 (6) Age (years) 30–39 16 (15.7) 40–49 56 (50.9) 50–59 25 (22.7) 60+ 13 (11.8) Number of rectal anastomoses performed per annum
![[Complications after rectal prolapse surgery].](/assets/img/hold.png)
