The Incidence of 30-Day Adverse Events After Colonoscopy
The Incidence of 30-Day Adverse Events After Colonoscopy
Major adverse events during colonoscopy occur in a small proportion of patients. Few data are available on the incidence of adverse events in the period after the procedure. Especially little is known about the occurrence of minor adverse events, which might influence the perceived burden of a colonoscopy and thereby the willingness to return.
This study supports that only a small proportion of patients experience major adverse events in the 30 days after their procedure (0.8% definite-related adverse events). However, almost a third of the patients do experience definite-related minor adverse events. These complaints do not result in significant more days lost from work. However, they do affect the willingness to return for colonoscopy (81 vs. 88%).
The credo 'primum non nocere' (first, do no harm) underlies all medical practice. Therefore, adverse events have been a core issue in quality assurance for colonoscopy. Several guidelines have been published to set targets for the occurrence of major adverse events. Large database studies have shown that the occurrence of death (<1 in 14,000), perforation (<1 in 1,200), and bleeding (<1 in 600) are generally low. Limitations of these studies might be the use of administrative databases and failure to follow patients for a longer period. Furthermore, many studies were performed in colorectal cancer screening trials, single-center settings, academic settings, or focused on single complications. The results of this cross-sectional quality assessment represent the daily clinical practice. The perforation rate (0.1%) and bleeding rate (0.4%) in our study were similar to other series and below the suggested standards (perforation: 0.2%; bleeding: 1%). Compared with a large prospective cohort study from the United States in which patients were called at 7 and 30 days post-procedure, numbers of definite- or possible-related major adverse events were higher in our study: 1.3 vs. 0.33%. This difference might be explained by the fact that in the study from the United States the cohort was restricted to patients who underwent colonoscopy for screening or surveillance and thus the included patients might be healthier and have less complicated procedures. Contradictory, we found that patients undergoing colonoscopy for screening or surveillance reported more often adverse events. This result might be explained by the fact that we included all reported adverse events, including minor adverse events. One may hypothesize that patients who do not undergo colonoscopy because of symptoms may sooner notice and report health problems in the period after the colonoscopy. At last, the expected and perceived symptoms after colonoscopy may be different between asymptomatic and symptomatic patients.
As the number of major adverse events was small, we were unable to identify significant risk factors for the occurrence of definite-related major adverse events. However, we did show that they were reported more often in older patients and female patients. Age has been proven to be a risk factor before. Results on gender as a risk factor are more inconsistent. A systematic review showed that female gender was a risk factor for colonic perforation (OR: 2.3). On the other hand, a Canadian population-based cohort study found a negative association between female gender and the risk of perforation or bleeding (OR: 0.67). However, for perforation alone, females did have an increased risk (OR: 1.21). In our study, the inclusion of abdominal discomfort might explain why female gender was found to be a risk factor for reporting major adverse events, as they generally experience more discomfort during and after colonoscopy.
Besides the well-known adverse events, in our study other major adverse events that made the patient visit the emergency department were observed as well. Some might be related to sedation or bowel preparation (angina and dizziness), others were probably related to the colonoscopic technique or interventions (discomfort). Furthermore, we observed more uncommon adverse events, such as transient ischemic attack, pulmonary embolism, and syncope. We could not prove a definite relation between the occurrence of these adverse events and the colonoscopy, therefore classified them as possible-related, as has been done before. However, the clinical impact of these adverse events will be the same, as patients will most probably link them to the colonoscopy. Our results thus show that the real incidence of adverse events will be underestimated both in number and in impact when only definitive adverse events are counted. It underlines that conservative quality assessment may underestimate this issue. The results further underline the importance of proper information provision, medication plans, and monitoring of patients after their colonoscopy.
Besides the major adverse events, patients may experience significant burden from minor adverse events. Previous research has found that these minor adverse events occur in 34% of patients undergoing a colonoscopy. Limitations of that study were the smaller sample size (n=470), the academic single-center setting of the study, no assessment of bowel complaints before the colonoscopy, and inclusion of screening and surveillance colonoscopies only. We addressed these items and our results are similar, with definite-related adverse events occurring in 29% of the patients. With respect to the severity of minor adverse events, we provided detail on the duration and experienced severity of the most common minor events, abdominal discomfort, and rectal blood loss. Our results show that rectal blood loss mostly was a minor problem, but abdominal discomfort could hold on for several days, and was experienced as more burdensome than before the procedure for the majority of patients that experienced it. Additionally, our results show that patients often miss a day from work after the colonoscopy due to the procedure. However, there was only a small difference in absenteeism between all patients and patients who reported direct-related adverse events (36 vs. 41%), indicating that the adverse events in general may not keep patients from their normal work. The number of minor adverse events reported by our and the American study from 2007 are significantly higher compared to a study performed in 1997 in the United States, where 17% of the patients experienced any adverse event. This may be explained by their less comprehensive interview-technique, and the inclusion of colonoscopies performed by experienced colonoscopists only. In our study, patients reported significantly more often adverse events when the colonoscopy was performed by a fellow. A retrospective chart audit found that perforation did occur more often when the endoscopist had performed <200 lower gastrointestinal endoscopies. Moreover, has it been established that a longer procedure time increases the risk of minor adverse events. Taken these results together indicate that the risk for any (major and minor) adverse event is higher if procedures are performed by less-experienced endoscopists.
Our study does have some limitations. First, as approximately a quarter of the patients were not reached by telephone, we could have missed some patients with serious adverse events. However, all patient records were checked for 30-day mortality. We assured ourselves that these patients were not deceased. However, they might have been admitted for adverse events at the time of the telephone interview. Another limitation might be recall-bias. As we called 30 days after the procedure, patients might have forgotten the exact details of their complaints. However, this would not be the case for serious adverse events, and may lead to an underestimation of the occurrence of minor adverse events.
In conclusion, this study supports that a colonoscopy is a safe procedure in daily clinical practice. Serious adverse events do occur in a small proportion of patients, however. Moreover, should the burden of the procedure in terms of minor adverse events not be ignored as a significant part of the patients undergoing colonoscopy do experience complaints in the days following their procedure and this decreases their willingness to return for colonoscopy. Patients and physicians should be aware that a colonoscopy is a burdensome procedure. Regular monitoring of the occurrence of both minor and major adverse events can help endoscopy departments improve the patient-centered care.
Discussion
Major adverse events during colonoscopy occur in a small proportion of patients. Few data are available on the incidence of adverse events in the period after the procedure. Especially little is known about the occurrence of minor adverse events, which might influence the perceived burden of a colonoscopy and thereby the willingness to return.
This study supports that only a small proportion of patients experience major adverse events in the 30 days after their procedure (0.8% definite-related adverse events). However, almost a third of the patients do experience definite-related minor adverse events. These complaints do not result in significant more days lost from work. However, they do affect the willingness to return for colonoscopy (81 vs. 88%).
The credo 'primum non nocere' (first, do no harm) underlies all medical practice. Therefore, adverse events have been a core issue in quality assurance for colonoscopy. Several guidelines have been published to set targets for the occurrence of major adverse events. Large database studies have shown that the occurrence of death (<1 in 14,000), perforation (<1 in 1,200), and bleeding (<1 in 600) are generally low. Limitations of these studies might be the use of administrative databases and failure to follow patients for a longer period. Furthermore, many studies were performed in colorectal cancer screening trials, single-center settings, academic settings, or focused on single complications. The results of this cross-sectional quality assessment represent the daily clinical practice. The perforation rate (0.1%) and bleeding rate (0.4%) in our study were similar to other series and below the suggested standards (perforation: 0.2%; bleeding: 1%). Compared with a large prospective cohort study from the United States in which patients were called at 7 and 30 days post-procedure, numbers of definite- or possible-related major adverse events were higher in our study: 1.3 vs. 0.33%. This difference might be explained by the fact that in the study from the United States the cohort was restricted to patients who underwent colonoscopy for screening or surveillance and thus the included patients might be healthier and have less complicated procedures. Contradictory, we found that patients undergoing colonoscopy for screening or surveillance reported more often adverse events. This result might be explained by the fact that we included all reported adverse events, including minor adverse events. One may hypothesize that patients who do not undergo colonoscopy because of symptoms may sooner notice and report health problems in the period after the colonoscopy. At last, the expected and perceived symptoms after colonoscopy may be different between asymptomatic and symptomatic patients.
As the number of major adverse events was small, we were unable to identify significant risk factors for the occurrence of definite-related major adverse events. However, we did show that they were reported more often in older patients and female patients. Age has been proven to be a risk factor before. Results on gender as a risk factor are more inconsistent. A systematic review showed that female gender was a risk factor for colonic perforation (OR: 2.3). On the other hand, a Canadian population-based cohort study found a negative association between female gender and the risk of perforation or bleeding (OR: 0.67). However, for perforation alone, females did have an increased risk (OR: 1.21). In our study, the inclusion of abdominal discomfort might explain why female gender was found to be a risk factor for reporting major adverse events, as they generally experience more discomfort during and after colonoscopy.
Besides the well-known adverse events, in our study other major adverse events that made the patient visit the emergency department were observed as well. Some might be related to sedation or bowel preparation (angina and dizziness), others were probably related to the colonoscopic technique or interventions (discomfort). Furthermore, we observed more uncommon adverse events, such as transient ischemic attack, pulmonary embolism, and syncope. We could not prove a definite relation between the occurrence of these adverse events and the colonoscopy, therefore classified them as possible-related, as has been done before. However, the clinical impact of these adverse events will be the same, as patients will most probably link them to the colonoscopy. Our results thus show that the real incidence of adverse events will be underestimated both in number and in impact when only definitive adverse events are counted. It underlines that conservative quality assessment may underestimate this issue. The results further underline the importance of proper information provision, medication plans, and monitoring of patients after their colonoscopy.
Besides the major adverse events, patients may experience significant burden from minor adverse events. Previous research has found that these minor adverse events occur in 34% of patients undergoing a colonoscopy. Limitations of that study were the smaller sample size (n=470), the academic single-center setting of the study, no assessment of bowel complaints before the colonoscopy, and inclusion of screening and surveillance colonoscopies only. We addressed these items and our results are similar, with definite-related adverse events occurring in 29% of the patients. With respect to the severity of minor adverse events, we provided detail on the duration and experienced severity of the most common minor events, abdominal discomfort, and rectal blood loss. Our results show that rectal blood loss mostly was a minor problem, but abdominal discomfort could hold on for several days, and was experienced as more burdensome than before the procedure for the majority of patients that experienced it. Additionally, our results show that patients often miss a day from work after the colonoscopy due to the procedure. However, there was only a small difference in absenteeism between all patients and patients who reported direct-related adverse events (36 vs. 41%), indicating that the adverse events in general may not keep patients from their normal work. The number of minor adverse events reported by our and the American study from 2007 are significantly higher compared to a study performed in 1997 in the United States, where 17% of the patients experienced any adverse event. This may be explained by their less comprehensive interview-technique, and the inclusion of colonoscopies performed by experienced colonoscopists only. In our study, patients reported significantly more often adverse events when the colonoscopy was performed by a fellow. A retrospective chart audit found that perforation did occur more often when the endoscopist had performed <200 lower gastrointestinal endoscopies. Moreover, has it been established that a longer procedure time increases the risk of minor adverse events. Taken these results together indicate that the risk for any (major and minor) adverse event is higher if procedures are performed by less-experienced endoscopists.
Our study does have some limitations. First, as approximately a quarter of the patients were not reached by telephone, we could have missed some patients with serious adverse events. However, all patient records were checked for 30-day mortality. We assured ourselves that these patients were not deceased. However, they might have been admitted for adverse events at the time of the telephone interview. Another limitation might be recall-bias. As we called 30 days after the procedure, patients might have forgotten the exact details of their complaints. However, this would not be the case for serious adverse events, and may lead to an underestimation of the occurrence of minor adverse events.
In conclusion, this study supports that a colonoscopy is a safe procedure in daily clinical practice. Serious adverse events do occur in a small proportion of patients, however. Moreover, should the burden of the procedure in terms of minor adverse events not be ignored as a significant part of the patients undergoing colonoscopy do experience complaints in the days following their procedure and this decreases their willingness to return for colonoscopy. Patients and physicians should be aware that a colonoscopy is a burdensome procedure. Regular monitoring of the occurrence of both minor and major adverse events can help endoscopy departments improve the patient-centered care.
Source...