Dig Dis Sci (2015) 60:1016–1023 DOI 10.1007/s10620-015-3639-3
ORIGINAL ARTICLE
The Impact of Chronic Opioid Use on Colonoscopy Outcomes Salman Nusrat1,2 • Sultan Mahmood1,2 • Hussein Bitar1,2 • William M. Tierney1,2 Klaus Bielefeldt3 • Mohammad F. Madhoun1,2
•
Received: 20 November 2014 / Accepted: 20 March 2015 / Published online: 31 March 2015 Ó Springer Science+Business Media New York 2015
Abstract Background Endoscopic procedures are frequently performed on patients chronically on opioids, raising concerns about the safety and efficacy of conventional sedation. Aims We hypothesized that chronic opioid use is associated with longer procedure times, higher dosages of sedation medications, and an increase in adverse effects. Methods This is a retrospective review from June 2012 to June 2013. Patients on chronic opioids (opioids use C12 weeks) were compared to randomly selected patients matched for age, race, and sex. Multivariate regression analysis was performed to identify factors that were independently predictive of longer procedure times. Results Patients on chronic opioids required higher doses of fentanyl (122.0 ± 45.3 vs. 105.8 ± 47.2 lg; P \ 0.0001) and midazolam (5.3 ± 5.3 vs. 4.4 ± 2 mg; P = 0.0037) and were more likely to receive diphenhydramine (42.8 vs. 22.6 %; P \ 0.001). The induction period (11.3 ± 8.8 vs. 7.5 ± 4.0 min), duration of procedure (39.1 ± 17.5 vs. 33.4 ± 14.1 min), and recovery times (38.7 ± 15.3 vs. 33.8 ± 12.1 min) were significantly longer for patients on chronic opioids. In the multivariate regression analysis, opioid use
& Salman Nusrat [email protected] 1
Section of Digestive Diseases, Department of Internal Medicine, University of Oklahoma Health Sciences Center, 920 Stanton Young Blvd. WP 1345, Oklahoma City, OK 73104, USA
2
Veterans Affair Medical Center, Oklahoma City, OK, USA
3
Gastroenterology, Hepatology and Nutrition, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213, USA
123
was an independent predictor of longer procedure duration (P \ 0.05). Hypotensive episodes did not differ between groups (2.8 vs. 2.7 %; P = 0.8). However, patients on chronic opioids experienced more pain (13.4 vs. 5.9 %; P 0.001) and hypertensive episodes (8.1 vs. 2.8 %; P 0.002). Conclusion Patients on chronic opioids represent a highrisk population with longer procedural times and more discomfort, despite higher dosages of sedative agents. Prospective studies are required to define the risks and benefits of more costly alternative sedation strategies for patients on chronic opioids. Keywords Sedation Procedural duration Adverse events Discomfort Bowel preparation
Introduction Colonoscopy is commonly used to treat and diagnose diseases of the colon and terminal ileum and is the test of choice for colon cancer screening in the USA [1, 2], where an estimated 15 million colonoscopies are performed annually [3]. Colonoscopies are safe, and serious adverse events are rare, reported in less than 0.1 % of procedures [4]. Cardiopulmonary adverse events account for up to 50 % of the morbidity and mortality associated with colonoscopies. These events are related to the sedation rather than the procedure itself [5–7]. Earlier studies suggested that adverse event rates are correlated with age, discomfort during colonoscopy, poor bowel preparation, history of anxiety, depression, diabetes, alcohol abuse, and opioid use [7–12]. About 10 % of US adults receive at least one prescription for opioids as part of their treatment for pain [13]. The rate of opioid prescriptions has doubled in the last 10 years
Dig Dis Sci (2015) 60:1016–1023
[14]. The number of patients experiencing chronic pain exceeds the number of patients with diabetes, heart disease, and cancer combined [15]. This trend is correlated with campaigns for more effective pain management and the aforementioned dramatic increase in opioid prescriptions [16, 17]. One might expect that many gastrointestinal endoscopic procedures are performed in patients taking prescription opioids [18]. Chronic opioid use is associated with higher rates of adverse events in patients undergoing orthopedic surgery [19]. The constipating effect of opioids may lead to inadequate bowel preparation, failed colonoscopy, and a more frequent need for repeat colonoscopy [9]. An earlier study that included ten patients on opioids failed to demonstrate that chronic opioid use was associated with difficult sedation [20]. In contrary, other studies have suggested that development of opioid tolerance may require higher dosages of sedatives and additional analgesic agents, which in turn increase the likelihood of adverse effects [21]. Appropriate sedation and pain control are important quality markers for procedural intervention and are also correlated with increased compliance [8]. Based on the above concerns, we hypothesized that chronic opioid use increases the rates of inadequate sedation, prolongs overall procedures, and is associated with more procedural adverse events. Using a retrospective case–control design, we determined the effect of chronic opioid use on (1) medication use to achieve appropriate sedation, (2) duration of induction, procedure, and recovery times, and (3) adverse events.
Methods The study was a retrospective case–control analysis of patients who underwent colonoscopies at the Oklahoma City Veterans Affair Medical Center (VAMC) between July 2012 and June 2013. The study was approved by the Institutional Review Board at the University of Oklahoma Health Sciences Center and the VAMC Research and Development Committee. Patients were identified by searching the billing record using the procedural codes for colonoscopy and the time constraints defined above. Chronic opioid use was operationally defined as at least 5 mg of morphine or its equivalent per day for 12 weeks or more, as indicated by prescriptions or medication use listed in the electronic medical records. This cohort of chronic opioid users was compared to a randomly selected control group, matched for age, sex, and race, who had not taken opioids for at least 3 months prior to their colonoscopy. Average daily opioid use was extracted based on the recorded or prescribed medication use and expressed as morphine
1017
equivalents [13]. For patients with flexible dosing schedules (‘‘as needed’’), we operationally defined the maximum possible dose as daily intake. Patients were excluded if they had a history of colon resection, if the procedure was performed for neoplasia screening in inflammatory bowel disease or emergently for active gastrointestinal bleeding, or if the procedure was aborted. We also abstracted demographic information, comorbidities, including diagnosis of depression or an anxiety disorder, current or prior alcohol abuse, medication use, and prior abdominal surgeries. Patients were classified as non-smokers, former smokers, or current smokers. A single dose of four liters of polyethylene glycol solution was used for bowel cleansing. All patients were asked to adhere to a clear liquid diet for 48 h prior to the colonoscopy. Procedures were performed by a team comprised of a gastroenterology fellow, a supervising physician, a medication nurse, and an instrumentation nurse, according to VAMC protocol. Intravenous fentanyl and midazolam were administered every 2–3 min in multiples of 25 mcg and 1 mg, respectively, until adequate sedation was achieved. Intravenous diphenhydramine was used as an adjunct sedative at the endoscopist’s discretion. In addition to providing more sedation, we used other techniques, such as changes in patient’s position, reduction of loop, water immersion, and deflation, to address patients’ discomfort during the colonoscopy. Blood pressure, heart rate, cardiac rhythm, oxygen saturation, and breaths per minute were monitored throughout the procedure and during recovery. Bowel preparation was quantified using the Boston Bowel Preparation Scale (BBPS; maximum score of 9 and a segmental score of 0–3). We reviewed the procedure report and the nurses’ electronic documentation to extract the type and dose of sedative, the induction time, procedural duration, and recovery time. Pain during the procedure was recorded by the nurses either as present or as absent. Type and number of adverse events were also recorded. We reviewed the procedural report and the nurses’ documentation to extract information about hypoxic episodes (oxygen saturation \89 %), hypotension (systolic BP \90 mmHg), and use of reversal agents like naloxone or flumazenil. We reviewed pathology results and recorded the number of adenomas removed. The primary endpoints were time required for induction, procedural duration, and recovery time. Secondary endpoints were appropriateness of sedation, numbers and dosages of sedatives required, safety as defined by the complication rate, adenoma detection rate, and bowel preparation scores. The primary endpoint used for power calculation in this study was the mean fentanyl dose used among the two groups (opioid vs. non-opioid). Assuming 275 subjects in each group and alpha value of \0.05, we will have 80 %
123
1018
Dig Dis Sci (2015) 60:1016–1023
Table 1 Basic characteristics
Opioid (n = 320)
Non-opioid (n = 320)
P value
Age (mean ± SD)
60.2 ± 8.9
61.0 ± 10.5
Male sex, n (%)
301 (94.1)
304 (95)
0.29 0.60
White race, n (%)
255 (79.7)
238 (74.6)
0.18
BMI (mean ± SD)
30.1 ± 6.1
30.1 ± 6.5
Smoking, n (%)
65 (32.2)
49 (15.3)
\0.001
0.95
Alcohol, n (%)
61 (19.1)
34 (10.6)
0.003
Junior fellow, n (%)
161 (50.3)
183 (57)
0.08
Comorbidities Diabetes, n (%)
87 (27.2)
85 (26.6)
Dementia, n (%)
0 (0)
3 (0.94)
0.79 0.08
Cirrhosis, n (%)
6 (1.9)
4 (1.3)
0.52
COPD, n (%)
78 (24.4)
67 (20.9)
0.29
Hypothyroidism, n (%)
34 (10.6)
19 (5.9)
0.03
Stroke, n (%)
11 (3.4)
9 (2.8)
0.65
HTN, n (%) OSA, n (%)
237 (74.3) 38 (11.9)
220 (68.8) 36 (10.9)
0.10 0.72
CAD, n (%)
38 (11.9)
40 (12.5)
0.81
Depression, n (%)
137 (42.8)
71 (22.2)
\0.0001
Anxiety, n (%)
59 (18.4)
45 (14.1)
0.13
Screening
41
41
0.07
FIT-positive
101
91
0.7
Symptomatic
79
57
0.08
Surveillance
99
113
0.08
Indication
power to detect at least 10 lg difference between the two groups. SAS software (SAS Institute, Cary, NC, USA) was used for data analyses. Continuous variables were reported as mean ± SD. Categorical variables were reported as percentages. Two-sided t tests were used to compare the means of continuous variables. Chi-square tests were used to compare the categorical variables, including the proportions of side effects. A P value \0.05 was considered statistically significant. Multiple linear regression analysis was performed by considering all independent variables which had at least modest correlation (P \ 0.1) with opioid use. Forward, backward, and stepwise multiple regression analyses were used to identify the independent variables that were most strongly associated with the procedure’s induction time, duration, and recovery.
Results Baseline Data A total of 1847 colonoscopies were performed during the study period. Our screening identified 320 chronic opioid users. These patients were matched with equal number of
123
randomly selected non-opioid-using controls who underwent colonoscopy during the same period. Age, gender, race, and BMI were not significantly different between the groups (Table 1). Except for depression and hypothyroidism, there was no significant difference in the presence of comorbidities (Table 1). We found that patients on chronic opioids were more likely to be smokers (32.2 vs. 15.3 %, P \ 0.001) and had a higher rate of alcohol (19.1 vs. 6.9 %, P 0.004) or illicit drug abuse (23.5 vs. 6.2 %) than controls. There was no significant difference in the indications for the colonoscopy, with occult blood-positive stools being the most common indication (31.5 % opioid vs. 30.1 % non-opioid, P = 0.7). The level of expertise of the primary endoscopist was not statistically different (P = 0.1) with junior fellows having performed 57 and 50.3 % of the colonoscopies under the supervision of an attending physician in the opioid group and non-opioid group, respectively (Table 1). The mean prescribed opioid dose was 47.4 ± 31.5 mg/day of morphine equivalents. Hydrocodone was the most commonly used opioid (Table 2). In majority (90 %) of the patients, the indication for opioid use was non-malignant chronic painful condition of which back pain was the most common (36.6 %). A review of medication listed in the medical record showed that patients on opioids were more
Dig Dis Sci (2015) 60:1016–1023 Table 2 Types of opioid used
1019
Types of opioid
N
Hydrocodone
142
Oxycodone
20
Hydromorphone
16
Morphine
13
Codeine
8
Methadone
113
Table 3 Medication use Medication Tricyclic antidepressant, n (%) SSRI, n (%)
Opioid
Non-opioid
P value
15 (4.7)
4 (1.3)
0.01
111 (34.7)
33 (10.3)
\0.0001 0.28
Antipsychotics, n (%)
9 (2.8)
5 (1.6)
Benzodiazepine, n (%)
55 (17.3)
14 (4.4)
0.0001
*Selective serotonin reuptake inhibitors
likely to be on selective serotonin reuptake inhibitors (SSRI; 34.7 vs. 10.3 %, P \ 0.0001), tricyclic antidepressants (TCA; 4.7 vs. 1.3 %, P = 0.01), and benzodiazepines (17.3 vs. 4.4 %, P = 0.0001; Table 3). Procedural Findings and Adverse Events We excluded 3.9 % patients (2.7 % secondary to poor preparation and 1.2 % because of inability to reach cecum) from the opioid group and 3.4 % patients (poor preparation in 2.1 and 1.1 % difficult insertion) from the control group, secondary to incomplete procedures. This rate was not significantly different between the groups (P = 0.6). The induction period (11.30 ± 8.8 vs. 7.5 ± 4.0 min, P \ 0.0001), duration of procedure (39.1 ± 17.5 vs. 33.4 ± 14.1 min, P \ 0.0001), and recovery times (38.7 ± 15.3 vs. 33.8 ± 12.1 min, P \ 0.0001) were significantly longer for patients on opioids (Table 4). Our analysis identified junior fellow, smoking, alcohol, depression, dementia, Table 4 Procedure’s related characteristics among those with opioid use versus those with no history of opioid use
hypothyroidism, TCA use, SSRI use, benzodiazepine use, opioid use, presence of adenoma, BBPS, and the use of diphenhydramine during procedure as variables having at least modest correlation (P \ 0.1) with opioid use. For induction period, duration of procedure, and the recovery time, all three model selection procedures (forward, backward, and stepwise) selected the same final models. In the final models, opioid use was noted as an independent predictor of longer duration of induction, procedure, and recovery times (please see Table 5 for complete list of independent predictors for induction period, duration of procedure, and recovery time). We stratified patients on chronic opioids into quartiles based on daily opioid use and compared the first quartile to the fourth quartile. Patients in the fourth quartile required higher amounts of both fentanyl and midazolam during colonoscopy and had longer induction period and recovery duration, but these differences were not significant (Table 6). BBPS scores were significantly lower in patients on chronic opioids (5.9 ± 1.3 vs. 6.2 ± 1.3, P = 0.01). The cumulative dosage of sedatives required was higher in opioid users (fentanyl: 122.1 ± 45.3 vs. 105.8 ± 47.2 lg, P \ 0.0001; midazolam: 5.3 ± 5.3 vs. 4.4 ± 2.0 mg, P = 0.0037). Patients on opioids were also more likely to receive diphenhydramine (42.8 vs. 23.1 %, P \ 0.0001) as an adjunct to their sedation (Table 4). The overall number of adverse events was low. Recorded adverse events were limited to reversible and relatively minor cardiovascular and hypoxic episodes. Comparing groups based on chronic opioid use did not show differences in transient hypotensive episodes: 2.8 vs. 2.7 %, P = 0.80; apnea: 0 vs. 0.3 %, P = 0.32; tachypnea: 0.3 vs. 0 %, P = 0.32; or hypoxia: 0.6 vs. 0 %, P = 0.16. However, patients on opioids experienced more pain (13.4 vs. 5.9 %, P = 0.001) and hypertensive episodes (8.1 vs. 2.8 %, P = 0.002). Our analysis noted that risk of any adverse event was higher with history of opioid use (OR 2.13, 95 % CI 1.34–3.39, P = 0.001; Table 7).
Opioid (n = 320)
Non-opioid (n = 320)
P value
Induction (mean ± SD)
11.3 ± 8.8
7.5 ± 4.0
\0.0001
Duration (mean ± SD)
39.1 ± 17.5
33.4 ± 14.1
\0.0001
Recovery (mean ± SD) Fentanyl dose (mean ± SD) Midazolam dose (mean ± SD) Benadryl use, n (%) BBPS (mean ± SD) Colonscopies with adenoma, n (%)
38.7 ± 15.3
33.8 ± 12.1
\0.0001
122.0 ± 45.3
105.8 ± 47.2
\0.0001
5.3 ± 5.3
4.4 ± 2.0
137 (42.8) 5.9 ± 1.3 141 (44.2)
74 (23.1) 6.2 ± 1.3 167 (52.4)
0.0037 \0.0001 0.01 0.039
123
1020
Dig Dis Sci (2015) 60:1016–1023
Table 5 Multivariate regression analysis for (a) induction time, (b) duration time, (c) recovery time Estimate
Error
TCA Alcohol
BBPS Opioids Junior fellow
95 % CI
P value
Junior fellow
0.98
0.56–1.69
0.93
Poor prep. (0 or 1)
0.001
0.00– [10
0.98
\.0001
Adenoma found
1.001
0.69–1.45
0.99
3.71
0.54 1.58
0.03
Dementia
0.001
0.00– [10
0.99
1.87
0.76
0.01
Hypothyroidism
0.8
0.36–1.80
0.59
Alcohol use (active)
1.04
0.49–2.21
0.92
4.74
1.24
0.0001
Smoking (active)
1.92
1.21–3.03
0.005
-1.17 5.36
0.47 1.24
0.0147 \0.0001
Depression
1.61
1.01–2.59
0.048
Antidepressant use
1.0
0.69–1.45
0.99
5.18
1.24
\0.0001
Benzodiazepine use
0.62
0.30–1.22
0.16
Opioid use
2.13
1.34–3.39
0.001
(c) Recovery time Opioid
4.48
1.26
\0.0001
Benadryl
2.86
1.20
0.0186
We examined the effects of opioid therapy on number of colonoscopies with at least one adenoma detected as a composite marker of appropriate bowel preparation and effective colonoscopy. This rate was high overall, with fewer patients in the opioid group (44.2 vs. 52.4 %, P 0.039), but a similar mean number of adenomas (1.01 ± 3.4 vs. 1.21 ± 3.4, P = 0.17).
Discussion Chronic opioid therapy is increasingly common and has reached rates of about 5 % of outpatients and 25 % of hospitalized individuals in the VA system [22, 23]. Consistent with this high prevalence of opioid use, 17.3 % of patients undergoing a colonoscopy at our center had taken opioids for at least 3 months prior to their procedure. Our results showed that chronic opioid use affected key outcomes and resulted in longer procedural times, higher dosages of sedatives, and an increased requirement for adjunct sedatives, compared with matched controls. The overall adverse event rate was low, reflecting the safety of colonoscopy under moderate sedation in this cohort of patients. Despite the known effects of chronic opioid use on gastrointestinal motility, we did not see an impact on
Table 6 Comparison between patients using lower opioid dose and higher opioid dose
procedure completion rates. However, a higher likelihood of minor and transient adverse events, lower BBPS, and lower polyp detection rates indicate that chronic opioid use prolongs the procedures, slightly compromises sedation effectiveness, and may also decrease their overall quality. For patients who are not at increased risk of colon cancer, screening is recommended beginning at age 50. Therefore, it is not surprising that the mean age of both of our groups was close to 60. Most patients were Caucasians. Racial percentages were close to national numbers [24]. Although we did not observe a significant difference in the indications for colonoscopy, patients on chronic opioids had more procedures performed for abdominal pain and changes in bowel habits. There was a trend toward fewer colonoscopies performed for screening, surveillance, and fecal immunochemical test (FIT)-positive stools in patients on chronic opioids compared with controls. A history of depression, chronic pain, and the constipating effects of opioid use likely contributed to these findings. Our results are consistent with previous studies showing that psychiatric disorders, smoking, and substance abuse disorders are more common in opioid users [25]. The mechanism underlying the relationship between the mood disorders and opioid use is not well defined. Since psychiatric disorders are associated with a higher prevalence of somatic symptoms, including pain [26], opioids may in
First quartile (n = 88) Induction (mean ± SD)
10.23 ± 6.0
Fourth upper quartile (n = 84) 11.5 ± 9.6
P value 0.909
Duration (mean ± SD)
37.6 ± 13.9
37.4 ± 16.1
0.285
Recovery (mean ± SD)
38.8 ± 15.3
41.0 ± 14.9
0.341
117.3 ± 45.3
125.2 ± 39.3
0.226
4.7 ± 1.9
5.1 ± 1.7
0.113
Fentanyl dose (mean ± SD) Midazolam dose (mean ± SD) Benadryl use, n (%)
123
OR
-3.46
(b) Duration time Adenomas
Variable
P value
(a) Induction time Opioid
Table 7 Factors predictive of adverse events during colonoscopy
35 (39.7)
45 (46.4)
0.378
Dig Dis Sci (2015) 60:1016–1023
some cases be used to treat poorly differentiated constellations of pain and emotional symptoms. Patient tolerance is important for the successful and safe completion of endoscopic procedures. Although endoscopic procedures are performed without sedation in some countries,[90 % of US procedures are performed with sedation [27]. Sedation reduces anxiety and pain and also results in increased compliance and willingness to undergo repeat procedures. Failure to achieve adequate sedation can result in patient discomfort, unsatisfactory examination, incomplete procedures, lost time, and financial loss [28–30]. Our results showed that patients on opioids required higher doses of fentanyl and midazolam. Patients in the opioid group were also more likely to receive diphenhydramine as an adjunct with more than 90 % of patients receiving it at the start of procedure based on anticipated difficult sedation rather than in response to inadequate sedation. The increased amount of sedatives used during colonoscopy can likely be explained by ‘‘tolerance or desensitization,’’ which involves uncoupling of G proteins, resulting in receptors that are still expressed on the surface of the cell, but are less responsive to opioids [31]. Another mechanism has been proposed, in which the number of opioid receptors is reduced via endocytosis of surface receptors [31]. Additionally, opioid-induced hyperalgesia and hyperesthesia have been described, characterized by an increase in pain sensitivity in chronic opioid abusers via spinal sensitization to glutamate and substance P [32]. Allodynia, in which pain is elicited by a normal non-painful stimulus, may also occur in chronic opioid abusers [32]. Patients on opioids were more likely to take benzodiazepines, SSRIs, and TCAs. Long-term use of these medications can further lead to tolerance to common sedatives and may explain the increased sedation requirements in this population [33]. As healthcare costs continue to rise, efficiency and quality of care have become major emphases [34]. Our results showed that all three procedure components were significantly longer in patients on chronic opioids. Procedure times depend on several factors, including number of adenomas, bowel preparation, operator skill, and the effectiveness of sedation during procedure [35–38]. Of these, the two that likely accounted for prolonged procedural duration in our study cohort are (1) poor bowel preparation due to opioid action on l receptors, which subsequently inhibits gastrointestinal propulsive activity [39, 40], and (2) difficulty in achieving adequate sedation secondary to tolerance. Factors other than chronic opioid use predictive of longer induction and recovery (i.e., TCA and diphenhydramine use) were also significantly more common in patients on chronic opioids. Adenoma detection rate (ADR) is the most widely used metric to assess quality of colonoscopy and has been
1021
correlated with long-term outcomes, including interval to colon cancer diagnosis [41]. Since we didn’t limit our calculations to screening colonoscopies and majority of the colonoscopies were diagnostic, it is not surprising that the adenoma detection rates were high in both groups. Interestingly, the adenoma detection rate was lower in patients on opioids, which was not explained by differences in the level of expertise of the endoscopist or indications for the procedures. Considering the case–control design, other biological variables (e.g., age) are unlikely. A protective effect of opioids on adenoma formation is similarly unlikely, raising questions about the potential impact of potentially subtle difference in pre-procedural preparation and/or the appropriateness of sedation as previously reported by others [42]. Although pulmonary and cardiovascular problems were diagnosed in our cohort, overall adverse events were rare [7]. Comorbid conditions did not significantly affect the likelihood of adverse effects during colonoscopy. However, our results were limited, as we did not correct for disease stage or severity or use a composite comorbidity index to address the more complex disease burden in individuals. Patients on chronic opioids had more pain and hypertensive episodes. The increased percentage of patients reporting pain may have been due to a combination of the following factors: inadequate sedation, longer procedure times, history of chronic pain, coexisting depression, or opioid-induced hyperalgesia [32, 43, 44]. Rates of apnea, hypoxia, tachypnea, hypotension, bradycardia, and tachycardia were similar between groups. However, the infrequent occurrence of adverse effects and their relatively mild and transient nature will likely require an even larger sample size to detect relevant differences. Colorectal cancer screening rates are increasing in the US. Screening practices have shifted to colonoscopy as the preferred approach [45, 46]. Searches for ways to increase efficiency and address delays due to slower induction led to the introduction of propofol as an endoscopy agent with faster onset and deeper sedation. The use of propofol rapidly grew after its introduction and concerns initially focused on the expenses faced by providers [46, 47]. However, propofol anesthesia typically requires a dedicated and more highly trained professional, such as a nurse anesthetist, significantly adding to the cost of the procedure [48]. Moreover, recent studies raised concerns about a higher likelihood of adverse events with deeper procedural sedation [49]. Concerns related to the associated costs and the potential for higher complication rates raise questions about the increasingly widespread adoption of such approaches. Our data are consistent with earlier studies [50, 51] and demonstrate some drawbacks of conventional sedation, guided and supervised by the endoscopist. However, they also show that overall completion
123
1022
Dig Dis Sci (2015) 60:1016–1023
rates are comparable and that even higher risk patients, i.e., individuals with chronic opioid use, can complete procedures without undue risk of significant adverse effects. Although our study is an unique exploration of the impact of chronic opioid use on colonoscopy outcomes, it has limitations. Since it is a single-center study involving veterans, our patient population was largely skewed toward Caucasian males. At our endoscopy unit, patients are instructed to take their routine medication on the morning of the procedure. However, the ‘‘as needed’’ nature of the dosing and retrospective study design does not allow us to ascertain and quantify medication use and dosing for the morning of the procedure. Additionally, we operationally defined medication dosage of agents taken on an ‘‘as needed’’ basis using the upper limit of the prescribed daily dose, thus possibly skewing data. Despite the retrospective study design, all efforts were made to include variables which could potentially affect outcomes, and regression analysis further helped in identifying independent predictors. Finally, given the rarity of serious adverse events, a larger sample size would be necessary to detect differences. Patients on chronic opioids represent a high-risk population with longer procedure times, high rates of procedural discomfort, poor bowel preparation, and lower adenoma detection rates. A multifaceted approach focusing on sedation and bowel cleansing can potentially improve these outcomes. Deep sedation with propofol or alternative adjuncts to sedation appears justified as a means to shorten procedure duration and improve patient satisfaction. However, prospective studies are required to define the true risks and benefits of these alternative sedation strategies before routine use of more costly alternatives can be recommended. Conflict of interest
None.
References 1. Minoli G, Meucci G, Bortoli A, et al. The ASGE guidelines for the appropriate use of colonoscopy in an open access system. Gastrointest Endosc. 2000;52:39–44. 2. Davila RE, Rajan E, Baron TH. ASGE guideline: colorectal cancer screening and surveillance. Gastrointest Endosc. 2006;63: 546–557. 3. Seeff LC, Richards TB, Shapiro JA, et al. How many endoscopies are performed for colorectal cancer screening? Results from CDC’s survey of endoscopic capacity. Gastroenterology. 2004;127:1670–1677. 4. Blero D, Devie`re J. Endoscopic complications—avoidance and management. Nat Rev Gastroenterol Hepatol. 2012;9:162–172. 5. Quine M, Bell G, McCloy R, Charlton J, Devlin H, Hopkins A. Prospective audit of upper gastrointestinal endoscopy in two regions of England: safety, staffing, and sedation methods. Gut. 1995;36:462–467. 6. Bowles C, Leicester R, Romaya C, Swarbrick E, Williams C, Epstein O. A prospective study of colonoscopy practice in the UK
123
7.
8.
9.
10.
11. 12.
13.
14.
15. 16. 17.
18.
19.
20.
21. 22.
23.
24. 25.
26. 27.
28.
today: are we adequately prepared for national colorectal cancer screening tomorrow? Gut. 2004;53:277–283. Nelson DB, McQuaid KR, Bond JH, Lieberman DA, Weiss DG, Johnston TK. Procedural success and complications of large-scale screening colonoscopy. Gastrointest Endosc. 2002;55:307–314. Harewood GC, Wiersema MJ, Melton LJ III. A prospective, controlled assessment of factors influencing acceptance of screening colonoscopy. Am J Gastroenterol. 2002;97:3186–3194. Verma S, Fogel J, Beyda DJ, Bernstein B, Notar-Francesco V, Mohanty SR. Chronic methadone use, poor bowel visualization and failed colonoscopy: a preliminary study. World J Gastroenterol WJG. 2012;18:4350. Taylor C, Schubert ML. Decreased efficacy of polyethylene glycol lavage solution (golytely) in the preparation of diabetic patients for outpatient colonoscopy: a prospective and blinded study. Am J Gastroenterol. 2001;96:710–714. Ure T, Dehghan K, Vernava A III, Longo W, Andrus C, Daniel G. Colonoscopy in the elderly. Surg Endosc. 1995;9:505–508. Chung Y, Han D, Yoo K-S, Park C. Patient factors predictive of pain and difficulty during sedation-free colonoscopy: a prospective study in Korea. Dig Liver Dis. 2007;39:872–876. Sullivan MD, Edlund MJ, Fan M-Y, DeVries A, Brennan Braden J, Martin BC. Trends in use of opioids for non-cancer pain conditions 2000–2005 in commercial and Medicaid insurance plans: the TROUP study. Pain. 2008;138:440–449. Daubresse M, Chang H-Y, Yu Y, et al. Ambulatory diagnosis and treatment of nonmalignant pain in the United States, 2000–2010. Med Care. 2013;51:870–878. Gaskin DJ, Richard P. The economic costs of pain in the United States. J Pain. 2012;13:715–724. Rockville M. Substance Abuse and Mental Health Services Administration. Center for Substance Abuse Treatment 2009. Research CoAP, Care, Research IoMCoAP, Medicine Io. Relieving pain in America: a blueprint for transforming prevention, care, education, and research. Washington: National Academies Press; 2011. Dunn KM, Saunders KW, Rutter CM, et al. Opioid prescriptions for chronic pain and overdose a cohort study. Ann Int Med. 2010;152:85–92. Pizzi LT, Toner R, Foley K, et al. Relationship between potential opioid-related adverse effects and hospital length of stay in patients receiving opioids after orthopedic surgery. Pharmacother J Hum Pharmacol Drug Therapy. 2012;32:502–514. Bal BS, Crowell MD, Kohli DR, et al. What factors are associated with the difficult-to-sedate endoscopy patient? Dig Dis Sci. 2012;57:2527–2534. Bryson EO. The anesthetic implications of illicit opioid abuse. Int Anesthesiol Clin. 2011;49:67–78. Seal KH, Shi Y, Cohen G, et al. Association of mental health disorders with prescription opioids and high-risk opioid use in US veterans of Iraq and Afghanistan. JAMA. 2012;307:940–947. Mosher HJ, Jiang L, Vaughan Sarrazin MS, Cram P, Kaboli PJ, Vander Weg MW. Prevalence and characteristics of hospitalized adults on chronic opioid therapy. J Hosp Med. 2014;9:82–87. QuickFacts A. US Census Bureau. 2010. Demyttenaere K, Bruffaerts R, Lee S, et al. Mental disorders among persons with chronic back or neck pain: results from the World Mental Health Surveys. Pain. 2007;129:332–342. Sullivan M, Katon W. Somatization: the path between distress and somatic symptoms. APS J. 1993;2:141–149. McQuaid KR, Laine L. A systematic review and meta-analysis of randomized, controlled trials of moderate sedation for routine endoscopic procedures. Gastrointest Endosc. 2008;67:910–923. Shah HA, Paszat LF, Saskin R, Stukel TA, Rabeneck L. Factors associated with incomplete colonoscopy: a population-based study. Gastroenterology. 2007;132:2297–2303.
Dig Dis Sci (2015) 60:1016–1023 29. Kim WH, Cho YJ, Park JY, Min PK, Kang JK, Park IS. Factors affecting insertion time and patient discomfort during colonoscopy. Gastrointest Endosc. 2000;52:600–605. 30. Radaelli F, Meucci G, Sgroi G, Minoli G. Technical performance of colonoscopy: the key role of sedation/analgesia and other quality indicators. Am J Gastroenterol. 2008;103:1122–1130. 31. Joseph EK, Reichling DB, Levine JD. Shared mechanisms for opioid tolerance and a transition to chronic pain. J Neurosci. 2010;30:4660–4666. 32. Angst MS, Clark JD. Opioid-induced hyperalgesia: a qualitative systematic review. Anesthesiology. 2006;104:570–587. 33. Michelini S, Cassano G, Frare F, Perugi G. Long-term use of benzodiazepines: tolerance, dependence and clinical problems in anxiety and mood disorders. Pharmacopsychiatry. 1996;29:127–134. 34. Leung WD, Gelrud A. Ushering a New Era in gastroenterology: the flexible gastroscope. Dig Dis Sci. 2014;59:12–15. 35. Lee J, McCallion K, Acheson A, Irwin S. A prospective randomised study comparing polyethylene glycol and sodium phosphate bowel cleansing solutions for colonoscopy. Ulster Med J. 1999;68:68. 36. Terruzzi V, Meucci G, Radaelli F, Terreni N, Minoli G. Routine versus ‘‘on demand’’ sedation and analgesia for colonoscopy: a prospective randomized controlled trial. Gastrointest Endosc. 2001;54:169–174. 37. Froehlich F, Wietlisbach V, Gonvers J-J, Burnand B, Vader J-P. Impact of colonic cleansing on quality and diagnostic yield of colonoscopy: the European panel of appropriateness of gastrointestinal endoscopy European multicenter study. Gastrointest Endosc. 2005;61:378–384. 38. Bernstein C, Thorn M, Monsees K, Spell R, O’Connor JB. A prospective study of factors that determine cecal intubation time at colonoscopy. Gastrointest Endosc. 2005;61:72–75. 39. Wood J, Galligan J. Function of opioids in the enteric nervous system. Neurogastroenterol Motil. 2004;16:17–28. 40. Epstein FH, Stein C. The control of pain in peripheral tissue by opioids. N Engl J Med. 1995;332:1685–1690. 41. Corley DA, Jensen CD, Marks AR, et al. Variation of adenoma prevalence by age, sex, race, and colon location in a large
1023
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
population: implications for screening and quality programs. Clin Gastroenterol Hepatol. 2013;11:172–180. Sherer EA, Imler TD, Imperiale TF. The effect of colonoscopy preparation quality on adenoma detection rates. Gastrointest Endosc. 2012;75:545–553. Glomsaker TB, Hoff G, Kvaløy JT, Søreide K, Aabakken L, Søreide JA. Patient-reported outcome measures after endoscopic retrograde cholangiopancreatography: a prospective, multicentre study. Scand J Gastroenterol. 2013;48:868–876. Holme O, Bretthauer M, de Lange T, et al. Risk stratification to predict pain during unsedated colonoscopy: results of a multicenter cohort study. Endoscopy. 2013;45:691–696. Sinicrope PS, Goode EL, Limburg PJ, et al. A population-based study of prevalence and adherence trends in average risk colorectal cancer screening, 1997 to 2008. Cancer Epidemiol Biomarkers Prev. 2012;21:347–350. Vargo JJ, Bramley T, Meyer K, Nightengale B. Practice efficiency and economics: the case for rapid recovery sedation agents for colonoscopy in a screening population. J Clin Gastroenterol. 2007;41:591–598. Aisenberg J, Brill JV, Ladabaum U, Cohen LB. Sedation for gastrointestinal endoscopy: new practices, new economics. Am J Gastroenterol. 2005;100:996–1000. Hassan C, Rex D, Cooper G, Benamouzig R. Endoscopist-directed propofol administration versus anesthesiologist assistance for colorectal cancer screening: a cost–effectiveness analysis. Endoscopy. 2012;44:456–464. Hoff G. Colonoscopy: sliding into the depths of sedation—is this what we want? Nat Rev Gastroenterol Hepatol. 2013;10:325– 326. Cohen LB, Wecsler JS, Gaetano JN, et al. Endoscopic sedation in the United States: results from a nationwide survey. Am J Gastroenterol. 2006;101:967–974. Koshy G, Nair S, Norkus EP, Hertan HI, Pitchumoni C. Propofol versus midazolam and meperidine for conscious sedation in GI endoscopy. Am J Gastroenterol. 2000;95:1476–1479.
123
ORIGINAL ARTICLE
The Impact of Chronic Opioid Use on Colonoscopy Outcomes Salman Nusrat1,2 • Sultan Mahmood1,2 • Hussein Bitar1,2 • William M. Tierney1,2 Klaus Bielefeldt3 • Mohammad F. Madhoun1,2
•
Received: 20 November 2014 / Accepted: 20 March 2015 / Published online: 31 March 2015 Ó Springer Science+Business Media New York 2015
Abstract Background Endoscopic procedures are frequently performed on patients chronically on opioids, raising concerns about the safety and efficacy of conventional sedation. Aims We hypothesized that chronic opioid use is associated with longer procedure times, higher dosages of sedation medications, and an increase in adverse effects. Methods This is a retrospective review from June 2012 to June 2013. Patients on chronic opioids (opioids use C12 weeks) were compared to randomly selected patients matched for age, race, and sex. Multivariate regression analysis was performed to identify factors that were independently predictive of longer procedure times. Results Patients on chronic opioids required higher doses of fentanyl (122.0 ± 45.3 vs. 105.8 ± 47.2 lg; P \ 0.0001) and midazolam (5.3 ± 5.3 vs. 4.4 ± 2 mg; P = 0.0037) and were more likely to receive diphenhydramine (42.8 vs. 22.6 %; P \ 0.001). The induction period (11.3 ± 8.8 vs. 7.5 ± 4.0 min), duration of procedure (39.1 ± 17.5 vs. 33.4 ± 14.1 min), and recovery times (38.7 ± 15.3 vs. 33.8 ± 12.1 min) were significantly longer for patients on chronic opioids. In the multivariate regression analysis, opioid use
& Salman Nusrat [email protected] 1
Section of Digestive Diseases, Department of Internal Medicine, University of Oklahoma Health Sciences Center, 920 Stanton Young Blvd. WP 1345, Oklahoma City, OK 73104, USA
2
Veterans Affair Medical Center, Oklahoma City, OK, USA
3
Gastroenterology, Hepatology and Nutrition, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213, USA
123
was an independent predictor of longer procedure duration (P \ 0.05). Hypotensive episodes did not differ between groups (2.8 vs. 2.7 %; P = 0.8). However, patients on chronic opioids experienced more pain (13.4 vs. 5.9 %; P 0.001) and hypertensive episodes (8.1 vs. 2.8 %; P 0.002). Conclusion Patients on chronic opioids represent a highrisk population with longer procedural times and more discomfort, despite higher dosages of sedative agents. Prospective studies are required to define the risks and benefits of more costly alternative sedation strategies for patients on chronic opioids. Keywords Sedation Procedural duration Adverse events Discomfort Bowel preparation
Introduction Colonoscopy is commonly used to treat and diagnose diseases of the colon and terminal ileum and is the test of choice for colon cancer screening in the USA [1, 2], where an estimated 15 million colonoscopies are performed annually [3]. Colonoscopies are safe, and serious adverse events are rare, reported in less than 0.1 % of procedures [4]. Cardiopulmonary adverse events account for up to 50 % of the morbidity and mortality associated with colonoscopies. These events are related to the sedation rather than the procedure itself [5–7]. Earlier studies suggested that adverse event rates are correlated with age, discomfort during colonoscopy, poor bowel preparation, history of anxiety, depression, diabetes, alcohol abuse, and opioid use [7–12]. About 10 % of US adults receive at least one prescription for opioids as part of their treatment for pain [13]. The rate of opioid prescriptions has doubled in the last 10 years
Dig Dis Sci (2015) 60:1016–1023
[14]. The number of patients experiencing chronic pain exceeds the number of patients with diabetes, heart disease, and cancer combined [15]. This trend is correlated with campaigns for more effective pain management and the aforementioned dramatic increase in opioid prescriptions [16, 17]. One might expect that many gastrointestinal endoscopic procedures are performed in patients taking prescription opioids [18]. Chronic opioid use is associated with higher rates of adverse events in patients undergoing orthopedic surgery [19]. The constipating effect of opioids may lead to inadequate bowel preparation, failed colonoscopy, and a more frequent need for repeat colonoscopy [9]. An earlier study that included ten patients on opioids failed to demonstrate that chronic opioid use was associated with difficult sedation [20]. In contrary, other studies have suggested that development of opioid tolerance may require higher dosages of sedatives and additional analgesic agents, which in turn increase the likelihood of adverse effects [21]. Appropriate sedation and pain control are important quality markers for procedural intervention and are also correlated with increased compliance [8]. Based on the above concerns, we hypothesized that chronic opioid use increases the rates of inadequate sedation, prolongs overall procedures, and is associated with more procedural adverse events. Using a retrospective case–control design, we determined the effect of chronic opioid use on (1) medication use to achieve appropriate sedation, (2) duration of induction, procedure, and recovery times, and (3) adverse events.
Methods The study was a retrospective case–control analysis of patients who underwent colonoscopies at the Oklahoma City Veterans Affair Medical Center (VAMC) between July 2012 and June 2013. The study was approved by the Institutional Review Board at the University of Oklahoma Health Sciences Center and the VAMC Research and Development Committee. Patients were identified by searching the billing record using the procedural codes for colonoscopy and the time constraints defined above. Chronic opioid use was operationally defined as at least 5 mg of morphine or its equivalent per day for 12 weeks or more, as indicated by prescriptions or medication use listed in the electronic medical records. This cohort of chronic opioid users was compared to a randomly selected control group, matched for age, sex, and race, who had not taken opioids for at least 3 months prior to their colonoscopy. Average daily opioid use was extracted based on the recorded or prescribed medication use and expressed as morphine
1017
equivalents [13]. For patients with flexible dosing schedules (‘‘as needed’’), we operationally defined the maximum possible dose as daily intake. Patients were excluded if they had a history of colon resection, if the procedure was performed for neoplasia screening in inflammatory bowel disease or emergently for active gastrointestinal bleeding, or if the procedure was aborted. We also abstracted demographic information, comorbidities, including diagnosis of depression or an anxiety disorder, current or prior alcohol abuse, medication use, and prior abdominal surgeries. Patients were classified as non-smokers, former smokers, or current smokers. A single dose of four liters of polyethylene glycol solution was used for bowel cleansing. All patients were asked to adhere to a clear liquid diet for 48 h prior to the colonoscopy. Procedures were performed by a team comprised of a gastroenterology fellow, a supervising physician, a medication nurse, and an instrumentation nurse, according to VAMC protocol. Intravenous fentanyl and midazolam were administered every 2–3 min in multiples of 25 mcg and 1 mg, respectively, until adequate sedation was achieved. Intravenous diphenhydramine was used as an adjunct sedative at the endoscopist’s discretion. In addition to providing more sedation, we used other techniques, such as changes in patient’s position, reduction of loop, water immersion, and deflation, to address patients’ discomfort during the colonoscopy. Blood pressure, heart rate, cardiac rhythm, oxygen saturation, and breaths per minute were monitored throughout the procedure and during recovery. Bowel preparation was quantified using the Boston Bowel Preparation Scale (BBPS; maximum score of 9 and a segmental score of 0–3). We reviewed the procedure report and the nurses’ electronic documentation to extract the type and dose of sedative, the induction time, procedural duration, and recovery time. Pain during the procedure was recorded by the nurses either as present or as absent. Type and number of adverse events were also recorded. We reviewed the procedural report and the nurses’ documentation to extract information about hypoxic episodes (oxygen saturation \89 %), hypotension (systolic BP \90 mmHg), and use of reversal agents like naloxone or flumazenil. We reviewed pathology results and recorded the number of adenomas removed. The primary endpoints were time required for induction, procedural duration, and recovery time. Secondary endpoints were appropriateness of sedation, numbers and dosages of sedatives required, safety as defined by the complication rate, adenoma detection rate, and bowel preparation scores. The primary endpoint used for power calculation in this study was the mean fentanyl dose used among the two groups (opioid vs. non-opioid). Assuming 275 subjects in each group and alpha value of \0.05, we will have 80 %
123
1018
Dig Dis Sci (2015) 60:1016–1023
Table 1 Basic characteristics
Opioid (n = 320)
Non-opioid (n = 320)
P value
Age (mean ± SD)
60.2 ± 8.9
61.0 ± 10.5
Male sex, n (%)
301 (94.1)
304 (95)
0.29 0.60
White race, n (%)
255 (79.7)
238 (74.6)
0.18
BMI (mean ± SD)
30.1 ± 6.1
30.1 ± 6.5
Smoking, n (%)
65 (32.2)
49 (15.3)
\0.001
0.95
Alcohol, n (%)
61 (19.1)
34 (10.6)
0.003
Junior fellow, n (%)
161 (50.3)
183 (57)
0.08
Comorbidities Diabetes, n (%)
87 (27.2)
85 (26.6)
Dementia, n (%)
0 (0)
3 (0.94)
0.79 0.08
Cirrhosis, n (%)
6 (1.9)
4 (1.3)
0.52
COPD, n (%)
78 (24.4)
67 (20.9)
0.29
Hypothyroidism, n (%)
34 (10.6)
19 (5.9)
0.03
Stroke, n (%)
11 (3.4)
9 (2.8)
0.65
HTN, n (%) OSA, n (%)
237 (74.3) 38 (11.9)
220 (68.8) 36 (10.9)
0.10 0.72
CAD, n (%)
38 (11.9)
40 (12.5)
0.81
Depression, n (%)
137 (42.8)
71 (22.2)
\0.0001
Anxiety, n (%)
59 (18.4)
45 (14.1)
0.13
Screening
41
41
0.07
FIT-positive
101
91
0.7
Symptomatic
79
57
0.08
Surveillance
99
113
0.08
Indication
power to detect at least 10 lg difference between the two groups. SAS software (SAS Institute, Cary, NC, USA) was used for data analyses. Continuous variables were reported as mean ± SD. Categorical variables were reported as percentages. Two-sided t tests were used to compare the means of continuous variables. Chi-square tests were used to compare the categorical variables, including the proportions of side effects. A P value \0.05 was considered statistically significant. Multiple linear regression analysis was performed by considering all independent variables which had at least modest correlation (P \ 0.1) with opioid use. Forward, backward, and stepwise multiple regression analyses were used to identify the independent variables that were most strongly associated with the procedure’s induction time, duration, and recovery.
Results Baseline Data A total of 1847 colonoscopies were performed during the study period. Our screening identified 320 chronic opioid users. These patients were matched with equal number of
123
randomly selected non-opioid-using controls who underwent colonoscopy during the same period. Age, gender, race, and BMI were not significantly different between the groups (Table 1). Except for depression and hypothyroidism, there was no significant difference in the presence of comorbidities (Table 1). We found that patients on chronic opioids were more likely to be smokers (32.2 vs. 15.3 %, P \ 0.001) and had a higher rate of alcohol (19.1 vs. 6.9 %, P 0.004) or illicit drug abuse (23.5 vs. 6.2 %) than controls. There was no significant difference in the indications for the colonoscopy, with occult blood-positive stools being the most common indication (31.5 % opioid vs. 30.1 % non-opioid, P = 0.7). The level of expertise of the primary endoscopist was not statistically different (P = 0.1) with junior fellows having performed 57 and 50.3 % of the colonoscopies under the supervision of an attending physician in the opioid group and non-opioid group, respectively (Table 1). The mean prescribed opioid dose was 47.4 ± 31.5 mg/day of morphine equivalents. Hydrocodone was the most commonly used opioid (Table 2). In majority (90 %) of the patients, the indication for opioid use was non-malignant chronic painful condition of which back pain was the most common (36.6 %). A review of medication listed in the medical record showed that patients on opioids were more
Dig Dis Sci (2015) 60:1016–1023 Table 2 Types of opioid used
1019
Types of opioid
N
Hydrocodone
142
Oxycodone
20
Hydromorphone
16
Morphine
13
Codeine
8
Methadone
113
Table 3 Medication use Medication Tricyclic antidepressant, n (%) SSRI, n (%)
Opioid
Non-opioid
P value
15 (4.7)
4 (1.3)
0.01
111 (34.7)
33 (10.3)
\0.0001 0.28
Antipsychotics, n (%)
9 (2.8)
5 (1.6)
Benzodiazepine, n (%)
55 (17.3)
14 (4.4)
0.0001
*Selective serotonin reuptake inhibitors
likely to be on selective serotonin reuptake inhibitors (SSRI; 34.7 vs. 10.3 %, P \ 0.0001), tricyclic antidepressants (TCA; 4.7 vs. 1.3 %, P = 0.01), and benzodiazepines (17.3 vs. 4.4 %, P = 0.0001; Table 3). Procedural Findings and Adverse Events We excluded 3.9 % patients (2.7 % secondary to poor preparation and 1.2 % because of inability to reach cecum) from the opioid group and 3.4 % patients (poor preparation in 2.1 and 1.1 % difficult insertion) from the control group, secondary to incomplete procedures. This rate was not significantly different between the groups (P = 0.6). The induction period (11.30 ± 8.8 vs. 7.5 ± 4.0 min, P \ 0.0001), duration of procedure (39.1 ± 17.5 vs. 33.4 ± 14.1 min, P \ 0.0001), and recovery times (38.7 ± 15.3 vs. 33.8 ± 12.1 min, P \ 0.0001) were significantly longer for patients on opioids (Table 4). Our analysis identified junior fellow, smoking, alcohol, depression, dementia, Table 4 Procedure’s related characteristics among those with opioid use versus those with no history of opioid use
hypothyroidism, TCA use, SSRI use, benzodiazepine use, opioid use, presence of adenoma, BBPS, and the use of diphenhydramine during procedure as variables having at least modest correlation (P \ 0.1) with opioid use. For induction period, duration of procedure, and the recovery time, all three model selection procedures (forward, backward, and stepwise) selected the same final models. In the final models, opioid use was noted as an independent predictor of longer duration of induction, procedure, and recovery times (please see Table 5 for complete list of independent predictors for induction period, duration of procedure, and recovery time). We stratified patients on chronic opioids into quartiles based on daily opioid use and compared the first quartile to the fourth quartile. Patients in the fourth quartile required higher amounts of both fentanyl and midazolam during colonoscopy and had longer induction period and recovery duration, but these differences were not significant (Table 6). BBPS scores were significantly lower in patients on chronic opioids (5.9 ± 1.3 vs. 6.2 ± 1.3, P = 0.01). The cumulative dosage of sedatives required was higher in opioid users (fentanyl: 122.1 ± 45.3 vs. 105.8 ± 47.2 lg, P \ 0.0001; midazolam: 5.3 ± 5.3 vs. 4.4 ± 2.0 mg, P = 0.0037). Patients on opioids were also more likely to receive diphenhydramine (42.8 vs. 23.1 %, P \ 0.0001) as an adjunct to their sedation (Table 4). The overall number of adverse events was low. Recorded adverse events were limited to reversible and relatively minor cardiovascular and hypoxic episodes. Comparing groups based on chronic opioid use did not show differences in transient hypotensive episodes: 2.8 vs. 2.7 %, P = 0.80; apnea: 0 vs. 0.3 %, P = 0.32; tachypnea: 0.3 vs. 0 %, P = 0.32; or hypoxia: 0.6 vs. 0 %, P = 0.16. However, patients on opioids experienced more pain (13.4 vs. 5.9 %, P = 0.001) and hypertensive episodes (8.1 vs. 2.8 %, P = 0.002). Our analysis noted that risk of any adverse event was higher with history of opioid use (OR 2.13, 95 % CI 1.34–3.39, P = 0.001; Table 7).
Opioid (n = 320)
Non-opioid (n = 320)
P value
Induction (mean ± SD)
11.3 ± 8.8
7.5 ± 4.0
\0.0001
Duration (mean ± SD)
39.1 ± 17.5
33.4 ± 14.1
\0.0001
Recovery (mean ± SD) Fentanyl dose (mean ± SD) Midazolam dose (mean ± SD) Benadryl use, n (%) BBPS (mean ± SD) Colonscopies with adenoma, n (%)
38.7 ± 15.3
33.8 ± 12.1
\0.0001
122.0 ± 45.3
105.8 ± 47.2
\0.0001
5.3 ± 5.3
4.4 ± 2.0
137 (42.8) 5.9 ± 1.3 141 (44.2)
74 (23.1) 6.2 ± 1.3 167 (52.4)
0.0037 \0.0001 0.01 0.039
123
1020
Dig Dis Sci (2015) 60:1016–1023
Table 5 Multivariate regression analysis for (a) induction time, (b) duration time, (c) recovery time Estimate
Error
TCA Alcohol
BBPS Opioids Junior fellow
95 % CI
P value
Junior fellow
0.98
0.56–1.69
0.93
Poor prep. (0 or 1)
0.001
0.00– [10
0.98
\.0001
Adenoma found
1.001
0.69–1.45
0.99
3.71
0.54 1.58
0.03
Dementia
0.001
0.00– [10
0.99
1.87
0.76
0.01
Hypothyroidism
0.8
0.36–1.80
0.59
Alcohol use (active)
1.04
0.49–2.21
0.92
4.74
1.24
0.0001
Smoking (active)
1.92
1.21–3.03
0.005
-1.17 5.36
0.47 1.24
0.0147 \0.0001
Depression
1.61
1.01–2.59
0.048
Antidepressant use
1.0
0.69–1.45
0.99
5.18
1.24
\0.0001
Benzodiazepine use
0.62
0.30–1.22
0.16
Opioid use
2.13
1.34–3.39
0.001
(c) Recovery time Opioid
4.48
1.26
\0.0001
Benadryl
2.86
1.20
0.0186
We examined the effects of opioid therapy on number of colonoscopies with at least one adenoma detected as a composite marker of appropriate bowel preparation and effective colonoscopy. This rate was high overall, with fewer patients in the opioid group (44.2 vs. 52.4 %, P 0.039), but a similar mean number of adenomas (1.01 ± 3.4 vs. 1.21 ± 3.4, P = 0.17).
Discussion Chronic opioid therapy is increasingly common and has reached rates of about 5 % of outpatients and 25 % of hospitalized individuals in the VA system [22, 23]. Consistent with this high prevalence of opioid use, 17.3 % of patients undergoing a colonoscopy at our center had taken opioids for at least 3 months prior to their procedure. Our results showed that chronic opioid use affected key outcomes and resulted in longer procedural times, higher dosages of sedatives, and an increased requirement for adjunct sedatives, compared with matched controls. The overall adverse event rate was low, reflecting the safety of colonoscopy under moderate sedation in this cohort of patients. Despite the known effects of chronic opioid use on gastrointestinal motility, we did not see an impact on
Table 6 Comparison between patients using lower opioid dose and higher opioid dose
procedure completion rates. However, a higher likelihood of minor and transient adverse events, lower BBPS, and lower polyp detection rates indicate that chronic opioid use prolongs the procedures, slightly compromises sedation effectiveness, and may also decrease their overall quality. For patients who are not at increased risk of colon cancer, screening is recommended beginning at age 50. Therefore, it is not surprising that the mean age of both of our groups was close to 60. Most patients were Caucasians. Racial percentages were close to national numbers [24]. Although we did not observe a significant difference in the indications for colonoscopy, patients on chronic opioids had more procedures performed for abdominal pain and changes in bowel habits. There was a trend toward fewer colonoscopies performed for screening, surveillance, and fecal immunochemical test (FIT)-positive stools in patients on chronic opioids compared with controls. A history of depression, chronic pain, and the constipating effects of opioid use likely contributed to these findings. Our results are consistent with previous studies showing that psychiatric disorders, smoking, and substance abuse disorders are more common in opioid users [25]. The mechanism underlying the relationship between the mood disorders and opioid use is not well defined. Since psychiatric disorders are associated with a higher prevalence of somatic symptoms, including pain [26], opioids may in
First quartile (n = 88) Induction (mean ± SD)
10.23 ± 6.0
Fourth upper quartile (n = 84) 11.5 ± 9.6
P value 0.909
Duration (mean ± SD)
37.6 ± 13.9
37.4 ± 16.1
0.285
Recovery (mean ± SD)
38.8 ± 15.3
41.0 ± 14.9
0.341
117.3 ± 45.3
125.2 ± 39.3
0.226
4.7 ± 1.9
5.1 ± 1.7
0.113
Fentanyl dose (mean ± SD) Midazolam dose (mean ± SD) Benadryl use, n (%)
123
OR
-3.46
(b) Duration time Adenomas
Variable
P value
(a) Induction time Opioid
Table 7 Factors predictive of adverse events during colonoscopy
35 (39.7)
45 (46.4)
0.378
Dig Dis Sci (2015) 60:1016–1023
some cases be used to treat poorly differentiated constellations of pain and emotional symptoms. Patient tolerance is important for the successful and safe completion of endoscopic procedures. Although endoscopic procedures are performed without sedation in some countries,[90 % of US procedures are performed with sedation [27]. Sedation reduces anxiety and pain and also results in increased compliance and willingness to undergo repeat procedures. Failure to achieve adequate sedation can result in patient discomfort, unsatisfactory examination, incomplete procedures, lost time, and financial loss [28–30]. Our results showed that patients on opioids required higher doses of fentanyl and midazolam. Patients in the opioid group were also more likely to receive diphenhydramine as an adjunct with more than 90 % of patients receiving it at the start of procedure based on anticipated difficult sedation rather than in response to inadequate sedation. The increased amount of sedatives used during colonoscopy can likely be explained by ‘‘tolerance or desensitization,’’ which involves uncoupling of G proteins, resulting in receptors that are still expressed on the surface of the cell, but are less responsive to opioids [31]. Another mechanism has been proposed, in which the number of opioid receptors is reduced via endocytosis of surface receptors [31]. Additionally, opioid-induced hyperalgesia and hyperesthesia have been described, characterized by an increase in pain sensitivity in chronic opioid abusers via spinal sensitization to glutamate and substance P [32]. Allodynia, in which pain is elicited by a normal non-painful stimulus, may also occur in chronic opioid abusers [32]. Patients on opioids were more likely to take benzodiazepines, SSRIs, and TCAs. Long-term use of these medications can further lead to tolerance to common sedatives and may explain the increased sedation requirements in this population [33]. As healthcare costs continue to rise, efficiency and quality of care have become major emphases [34]. Our results showed that all three procedure components were significantly longer in patients on chronic opioids. Procedure times depend on several factors, including number of adenomas, bowel preparation, operator skill, and the effectiveness of sedation during procedure [35–38]. Of these, the two that likely accounted for prolonged procedural duration in our study cohort are (1) poor bowel preparation due to opioid action on l receptors, which subsequently inhibits gastrointestinal propulsive activity [39, 40], and (2) difficulty in achieving adequate sedation secondary to tolerance. Factors other than chronic opioid use predictive of longer induction and recovery (i.e., TCA and diphenhydramine use) were also significantly more common in patients on chronic opioids. Adenoma detection rate (ADR) is the most widely used metric to assess quality of colonoscopy and has been
1021
correlated with long-term outcomes, including interval to colon cancer diagnosis [41]. Since we didn’t limit our calculations to screening colonoscopies and majority of the colonoscopies were diagnostic, it is not surprising that the adenoma detection rates were high in both groups. Interestingly, the adenoma detection rate was lower in patients on opioids, which was not explained by differences in the level of expertise of the endoscopist or indications for the procedures. Considering the case–control design, other biological variables (e.g., age) are unlikely. A protective effect of opioids on adenoma formation is similarly unlikely, raising questions about the potential impact of potentially subtle difference in pre-procedural preparation and/or the appropriateness of sedation as previously reported by others [42]. Although pulmonary and cardiovascular problems were diagnosed in our cohort, overall adverse events were rare [7]. Comorbid conditions did not significantly affect the likelihood of adverse effects during colonoscopy. However, our results were limited, as we did not correct for disease stage or severity or use a composite comorbidity index to address the more complex disease burden in individuals. Patients on chronic opioids had more pain and hypertensive episodes. The increased percentage of patients reporting pain may have been due to a combination of the following factors: inadequate sedation, longer procedure times, history of chronic pain, coexisting depression, or opioid-induced hyperalgesia [32, 43, 44]. Rates of apnea, hypoxia, tachypnea, hypotension, bradycardia, and tachycardia were similar between groups. However, the infrequent occurrence of adverse effects and their relatively mild and transient nature will likely require an even larger sample size to detect relevant differences. Colorectal cancer screening rates are increasing in the US. Screening practices have shifted to colonoscopy as the preferred approach [45, 46]. Searches for ways to increase efficiency and address delays due to slower induction led to the introduction of propofol as an endoscopy agent with faster onset and deeper sedation. The use of propofol rapidly grew after its introduction and concerns initially focused on the expenses faced by providers [46, 47]. However, propofol anesthesia typically requires a dedicated and more highly trained professional, such as a nurse anesthetist, significantly adding to the cost of the procedure [48]. Moreover, recent studies raised concerns about a higher likelihood of adverse events with deeper procedural sedation [49]. Concerns related to the associated costs and the potential for higher complication rates raise questions about the increasingly widespread adoption of such approaches. Our data are consistent with earlier studies [50, 51] and demonstrate some drawbacks of conventional sedation, guided and supervised by the endoscopist. However, they also show that overall completion
123
1022
Dig Dis Sci (2015) 60:1016–1023
rates are comparable and that even higher risk patients, i.e., individuals with chronic opioid use, can complete procedures without undue risk of significant adverse effects. Although our study is an unique exploration of the impact of chronic opioid use on colonoscopy outcomes, it has limitations. Since it is a single-center study involving veterans, our patient population was largely skewed toward Caucasian males. At our endoscopy unit, patients are instructed to take their routine medication on the morning of the procedure. However, the ‘‘as needed’’ nature of the dosing and retrospective study design does not allow us to ascertain and quantify medication use and dosing for the morning of the procedure. Additionally, we operationally defined medication dosage of agents taken on an ‘‘as needed’’ basis using the upper limit of the prescribed daily dose, thus possibly skewing data. Despite the retrospective study design, all efforts were made to include variables which could potentially affect outcomes, and regression analysis further helped in identifying independent predictors. Finally, given the rarity of serious adverse events, a larger sample size would be necessary to detect differences. Patients on chronic opioids represent a high-risk population with longer procedure times, high rates of procedural discomfort, poor bowel preparation, and lower adenoma detection rates. A multifaceted approach focusing on sedation and bowel cleansing can potentially improve these outcomes. Deep sedation with propofol or alternative adjuncts to sedation appears justified as a means to shorten procedure duration and improve patient satisfaction. However, prospective studies are required to define the true risks and benefits of these alternative sedation strategies before routine use of more costly alternatives can be recommended. Conflict of interest
None.
References 1. Minoli G, Meucci G, Bortoli A, et al. The ASGE guidelines for the appropriate use of colonoscopy in an open access system. Gastrointest Endosc. 2000;52:39–44. 2. Davila RE, Rajan E, Baron TH. ASGE guideline: colorectal cancer screening and surveillance. Gastrointest Endosc. 2006;63: 546–557. 3. Seeff LC, Richards TB, Shapiro JA, et al. How many endoscopies are performed for colorectal cancer screening? Results from CDC’s survey of endoscopic capacity. Gastroenterology. 2004;127:1670–1677. 4. Blero D, Devie`re J. Endoscopic complications—avoidance and management. Nat Rev Gastroenterol Hepatol. 2012;9:162–172. 5. Quine M, Bell G, McCloy R, Charlton J, Devlin H, Hopkins A. Prospective audit of upper gastrointestinal endoscopy in two regions of England: safety, staffing, and sedation methods. Gut. 1995;36:462–467. 6. Bowles C, Leicester R, Romaya C, Swarbrick E, Williams C, Epstein O. A prospective study of colonoscopy practice in the UK
123
7.
8.
9.
10.
11. 12.
13.
14.
15. 16. 17.
18.
19.
20.
21. 22.
23.
24. 25.
26. 27.
28.
today: are we adequately prepared for national colorectal cancer screening tomorrow? Gut. 2004;53:277–283. Nelson DB, McQuaid KR, Bond JH, Lieberman DA, Weiss DG, Johnston TK. Procedural success and complications of large-scale screening colonoscopy. Gastrointest Endosc. 2002;55:307–314. Harewood GC, Wiersema MJ, Melton LJ III. A prospective, controlled assessment of factors influencing acceptance of screening colonoscopy. Am J Gastroenterol. 2002;97:3186–3194. Verma S, Fogel J, Beyda DJ, Bernstein B, Notar-Francesco V, Mohanty SR. Chronic methadone use, poor bowel visualization and failed colonoscopy: a preliminary study. World J Gastroenterol WJG. 2012;18:4350. Taylor C, Schubert ML. Decreased efficacy of polyethylene glycol lavage solution (golytely) in the preparation of diabetic patients for outpatient colonoscopy: a prospective and blinded study. Am J Gastroenterol. 2001;96:710–714. Ure T, Dehghan K, Vernava A III, Longo W, Andrus C, Daniel G. Colonoscopy in the elderly. Surg Endosc. 1995;9:505–508. Chung Y, Han D, Yoo K-S, Park C. Patient factors predictive of pain and difficulty during sedation-free colonoscopy: a prospective study in Korea. Dig Liver Dis. 2007;39:872–876. Sullivan MD, Edlund MJ, Fan M-Y, DeVries A, Brennan Braden J, Martin BC. Trends in use of opioids for non-cancer pain conditions 2000–2005 in commercial and Medicaid insurance plans: the TROUP study. Pain. 2008;138:440–449. Daubresse M, Chang H-Y, Yu Y, et al. Ambulatory diagnosis and treatment of nonmalignant pain in the United States, 2000–2010. Med Care. 2013;51:870–878. Gaskin DJ, Richard P. The economic costs of pain in the United States. J Pain. 2012;13:715–724. Rockville M. Substance Abuse and Mental Health Services Administration. Center for Substance Abuse Treatment 2009. Research CoAP, Care, Research IoMCoAP, Medicine Io. Relieving pain in America: a blueprint for transforming prevention, care, education, and research. Washington: National Academies Press; 2011. Dunn KM, Saunders KW, Rutter CM, et al. Opioid prescriptions for chronic pain and overdose a cohort study. Ann Int Med. 2010;152:85–92. Pizzi LT, Toner R, Foley K, et al. Relationship between potential opioid-related adverse effects and hospital length of stay in patients receiving opioids after orthopedic surgery. Pharmacother J Hum Pharmacol Drug Therapy. 2012;32:502–514. Bal BS, Crowell MD, Kohli DR, et al. What factors are associated with the difficult-to-sedate endoscopy patient? Dig Dis Sci. 2012;57:2527–2534. Bryson EO. The anesthetic implications of illicit opioid abuse. Int Anesthesiol Clin. 2011;49:67–78. Seal KH, Shi Y, Cohen G, et al. Association of mental health disorders with prescription opioids and high-risk opioid use in US veterans of Iraq and Afghanistan. JAMA. 2012;307:940–947. Mosher HJ, Jiang L, Vaughan Sarrazin MS, Cram P, Kaboli PJ, Vander Weg MW. Prevalence and characteristics of hospitalized adults on chronic opioid therapy. J Hosp Med. 2014;9:82–87. QuickFacts A. US Census Bureau. 2010. Demyttenaere K, Bruffaerts R, Lee S, et al. Mental disorders among persons with chronic back or neck pain: results from the World Mental Health Surveys. Pain. 2007;129:332–342. Sullivan M, Katon W. Somatization: the path between distress and somatic symptoms. APS J. 1993;2:141–149. McQuaid KR, Laine L. A systematic review and meta-analysis of randomized, controlled trials of moderate sedation for routine endoscopic procedures. Gastrointest Endosc. 2008;67:910–923. Shah HA, Paszat LF, Saskin R, Stukel TA, Rabeneck L. Factors associated with incomplete colonoscopy: a population-based study. Gastroenterology. 2007;132:2297–2303.
Dig Dis Sci (2015) 60:1016–1023 29. Kim WH, Cho YJ, Park JY, Min PK, Kang JK, Park IS. Factors affecting insertion time and patient discomfort during colonoscopy. Gastrointest Endosc. 2000;52:600–605. 30. Radaelli F, Meucci G, Sgroi G, Minoli G. Technical performance of colonoscopy: the key role of sedation/analgesia and other quality indicators. Am J Gastroenterol. 2008;103:1122–1130. 31. Joseph EK, Reichling DB, Levine JD. Shared mechanisms for opioid tolerance and a transition to chronic pain. J Neurosci. 2010;30:4660–4666. 32. Angst MS, Clark JD. Opioid-induced hyperalgesia: a qualitative systematic review. Anesthesiology. 2006;104:570–587. 33. Michelini S, Cassano G, Frare F, Perugi G. Long-term use of benzodiazepines: tolerance, dependence and clinical problems in anxiety and mood disorders. Pharmacopsychiatry. 1996;29:127–134. 34. Leung WD, Gelrud A. Ushering a New Era in gastroenterology: the flexible gastroscope. Dig Dis Sci. 2014;59:12–15. 35. Lee J, McCallion K, Acheson A, Irwin S. A prospective randomised study comparing polyethylene glycol and sodium phosphate bowel cleansing solutions for colonoscopy. Ulster Med J. 1999;68:68. 36. Terruzzi V, Meucci G, Radaelli F, Terreni N, Minoli G. Routine versus ‘‘on demand’’ sedation and analgesia for colonoscopy: a prospective randomized controlled trial. Gastrointest Endosc. 2001;54:169–174. 37. Froehlich F, Wietlisbach V, Gonvers J-J, Burnand B, Vader J-P. Impact of colonic cleansing on quality and diagnostic yield of colonoscopy: the European panel of appropriateness of gastrointestinal endoscopy European multicenter study. Gastrointest Endosc. 2005;61:378–384. 38. Bernstein C, Thorn M, Monsees K, Spell R, O’Connor JB. A prospective study of factors that determine cecal intubation time at colonoscopy. Gastrointest Endosc. 2005;61:72–75. 39. Wood J, Galligan J. Function of opioids in the enteric nervous system. Neurogastroenterol Motil. 2004;16:17–28. 40. Epstein FH, Stein C. The control of pain in peripheral tissue by opioids. N Engl J Med. 1995;332:1685–1690. 41. Corley DA, Jensen CD, Marks AR, et al. Variation of adenoma prevalence by age, sex, race, and colon location in a large
1023
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
population: implications for screening and quality programs. Clin Gastroenterol Hepatol. 2013;11:172–180. Sherer EA, Imler TD, Imperiale TF. The effect of colonoscopy preparation quality on adenoma detection rates. Gastrointest Endosc. 2012;75:545–553. Glomsaker TB, Hoff G, Kvaløy JT, Søreide K, Aabakken L, Søreide JA. Patient-reported outcome measures after endoscopic retrograde cholangiopancreatography: a prospective, multicentre study. Scand J Gastroenterol. 2013;48:868–876. Holme O, Bretthauer M, de Lange T, et al. Risk stratification to predict pain during unsedated colonoscopy: results of a multicenter cohort study. Endoscopy. 2013;45:691–696. Sinicrope PS, Goode EL, Limburg PJ, et al. A population-based study of prevalence and adherence trends in average risk colorectal cancer screening, 1997 to 2008. Cancer Epidemiol Biomarkers Prev. 2012;21:347–350. Vargo JJ, Bramley T, Meyer K, Nightengale B. Practice efficiency and economics: the case for rapid recovery sedation agents for colonoscopy in a screening population. J Clin Gastroenterol. 2007;41:591–598. Aisenberg J, Brill JV, Ladabaum U, Cohen LB. Sedation for gastrointestinal endoscopy: new practices, new economics. Am J Gastroenterol. 2005;100:996–1000. Hassan C, Rex D, Cooper G, Benamouzig R. Endoscopist-directed propofol administration versus anesthesiologist assistance for colorectal cancer screening: a cost–effectiveness analysis. Endoscopy. 2012;44:456–464. Hoff G. Colonoscopy: sliding into the depths of sedation—is this what we want? Nat Rev Gastroenterol Hepatol. 2013;10:325– 326. Cohen LB, Wecsler JS, Gaetano JN, et al. Endoscopic sedation in the United States: results from a nationwide survey. Am J Gastroenterol. 2006;101:967–974. Koshy G, Nair S, Norkus EP, Hertan HI, Pitchumoni C. Propofol versus midazolam and meperidine for conscious sedation in GI endoscopy. Am J Gastroenterol. 2000;95:1476–1479.
123
