Improved tolerability and efficaciousness of MiraLAX-Gatorade when compared to GoLYTELY in the outpatient setting

A clinical decision report using Gu P, Lew D, Jung OS, et al. Comparing the Real-World Effectiveness of Competing Colonoscopy Preparations: Results of a Prospective Trial. Am J Gastroenterol. 2019;114(2):305-314. https://doi.org/10.14309/ajg.0000000000000057 for a patient reluctant to undergo colonoscopy screening due to concerns about GoLYTLEY's tolerability.

Improved tolerability and efficaciousness of MiraLAX-Gatorade when compared Improved tolerability and efficaciousness of MiraLAX-Gatorade when compared to GoLYTELY in the outpatient setting to GoLYTELY in the outpatient setting Cover Page Footnote Cover Page Footnote I would like to thank Dr. Susan Stevens and Dr. Tamera Schmidt for allowing me to rotate through their clinic.

Related Literature
A search on UpToDate for "bowel preparation" provided a list of the FDA approved colonoscopy preparations. This included 4L polyethylene glycol with electrolyte solution (PEG-ELS), which includes GoLYTELY, 2L PEG-ELS, which includes Moviprep, and hyperosmotic agents such as Suprep. MiraLAX is a non-FDA approved OTC bowel preparation that works similarly to GoLYTELY. It is often mixed with 2L of Gatorade and used as bowel preparation. A PubMed search of "colonoscopy bowel preparation AND Gatorade" and colonoscopy bowel preparation AND MiraLAX provided several articles on this topic. There is conflicting literature on the bowel cleansing effectiveness of MiraLAX-Gatorade when compared to GoLYTELY. The Enestvedt and the Hjelkrem studies were among the first randomized controlled trials (RCTs) to study to efficacy of MiraLAX-Gatorade. They both concluded that it was inferior to GoLYTELY in bowel preparation quality, determined by the Boston Bowel Preparation Scale (BBPS) scores and Ottawa scores respectively. 1,2 Another Enestvedt study later determined that MiraLAX-Gatorade also produced lower adenoma detection rates than GoLYTELY. 3 This led to the conclusion that GoLYTELY should be used as first line for bowel preparation.
There were 2 RCTs that concluded MiraLAX-Gatorade was a non-inferior preparation when compared to GoLYTELY. The different result in the McKenna study was attributed to the lack of split dosing, which was used prior. The McKenna study also had independent researchers assign the BBPS score based off pictures of bowel, as opposed to by the endoscopists. 4 The Samarasena study utilized the Ottawa score and the BBPS score, which was assigned by independent researchers observing a video of the procedure. In addition, MiraLAX-Gatorade consistently scored higher on patient tolerability surveys in this study, which asks about specific symptoms such as nausea, bloating, cramping, discomfort, and sleep disturbance. 5 This was the first study where bowel preparation tolerability was assessed. Another article was a retrospective endoscopic database analysis at a community hospital, which concluded that MiraLAX-Gatorade-bisacodyl provided a statistically significant improvement to bowel cleansing. 6 This was attributed to improved compliance when compared to GoLYTLEY. The Govani study, which was a retrospective analysis, tested the impact of many different variables on bowel preparation efficacy. It found that MiraLAX-Gatorade preparations were associated with better bowel preparation, assessed by endoscopists. 7 The first meta-analysis on MiraLAX-Gatorade effectiveness found that it was inferior to GoLYTLEY. 8 This analysis included 1418 patients and utilized 5 different studies, including the Enestvedt and Hjelkrem studies. A later meta-analysis by Zhang utilized regression with random effects on the same data and found that no significant difference existed between the groups, leading to the conclusion that many variables impact preparation efficacy. 9 Many of the previously listed studies tested the effectiveness of MiraLAX-Gatorade in a controlled setting with monitored protocols; however, the clinical effectiveness of MiraLAX-Gatorade in a "real-world" setting has not been extensively studied. Gu et al. provided a prospective, naturalistic comparative effectiveness study to assess the effectiveness and tolerability of different colonoscopy preparations. 10 This makes it more applicable to the outpatient setting, as patient tolerability can determine preparation effectiveness and was Mrs. Smith's primary concern. The level of evidence of this study is 2 using the SORT algorithm, as it is an observational study. Despite this, a study that mimicked outpatient colonoscopies and measured both efficacy and tolerability was required to answer the clinical question, making this paper ideal.

Critical Appraisal
This study was conducted at Cedars-Sinai Hospital in Los Angeles, California. To be eligible, patients had to have an outpatient colonoscopy from August 4th, 2016 to July 31st, 2018 and be ≥ 18. Age, BMI, race/ethnicity, sex, language, opioid/tricyclic antidepressant use, and comorbidities of patients were noted. Patients were primarily Non-Hispanic white (2952) and 51% female, with the majority aged 50-64 (1763). Bowel cleansing was assessed using the routine BBPS score, which is assigned by a nurse during the colonoscopy. Adequate bowel cleansing was defined as a total BBPS score of ≥ 6. Bowel preparation tolerability was assessed through a question asked to each patient. Prior to the procedure, the nurse asked each patient "How much of the bowel prep did you ingest," with answer options being all (100%), 75-99%, <75%, or unsure. This was further categorized into "fully completed" and "not fully completed" groups. Endoscopists were free to give any preparation and dose, and patients were asked which preparation they used prior to the procedure. GoLYTELY, Colyte, NuLYTELY, Trilyte, MoviPrep, MiraLAX-Gatorade, Suprep, Suclear, Prepopik, OsmoPrep, and magnesium citrate were all used; however, Colyte, NuLYTELY, and Trilyte were grouped with GoLYTELY due to similar formulation.
Overall, 5253 outpatient colonoscopies were completed, with 914 cases excluded either due to age < 18 (257), using two day preparations or unknown dosing (184), absent colons or procedure termination other than suboptimal preparation (93), or missing scoring data (383). The study included 4339 colonoscopies, performed by 75 different independent providers. The results showed that MiraLAX-Gatorade had a 92.6% completion rate, whereas GoLYTELY had an 82.9% completion rate. Upon adjusting for preparation, provider, and patient-related variables using multivariable logistic regression analysis with random effects, patients using MiraLAX were significantly more likely to finish the preparation, with a rate of 62.3% compared to 11.8% for GoLYTELY (p < 0.001). MiraLAX-Gatorade also had the highest adjusted completion rate when compared to the other preparations. With regards to the bowel cleansing efficacy, MiraLAX-Gatorade had an average BBPS score of 7.09±1.64 whereas GoLYTELY had an average score of 6.67±1.87 after multivariable regression analysis, which was significant (p < 0.001). Patients using MiraLAX-Gatorade also had higher odds for adequate cleansing when compared to GoLYTELY (OR, 1.76; 95% CI, 1.24-2.49). MoviPrep and Suprep also had higher completion rates and BBPS scores than GoLYTELY. Patients told to use split-prep dosing also had higher total BBPS scores.
Several characteristics contributed to the strength of this study. This was the largest prospective study analyzing both the tolerability and efficacy of various bowel preparations, with over 4300 colonoscopies included. In addition, because there were over 75 providers that could prescribe any regimen, this study can compare many different bowel preparations. Patients were also not excluded for having chronic conditions such as chronic opioid use, inflammatory bowel disease, chronic kidney disease, congestive heart failure, and cirrhosis. 10 This is in contrast to the RCTs previously listed, which typically include fewer, more controlled regimens and excluded patients with chronic conditions. 1,4,5 This study was also free of the Hawthorne effect and remuneration, as it was not a RCT. Multiple RCTs have found GoLYTELY to have superior efficacy when compared to MiraLAX-Gatorade 1,2,3 ; however, patients may be more willing to fully ingest GoLYTELY, despite its lower tolerability, due to these biases. This makes prospective, naturalistic trials like the Ru study more closely reflect outpatient clinical care. Regression analysis with random effects also improves the strength of these results, as it controlled for unobserved heterogeneity.
The authors chose to exclude patients less than 18 years old, those with surgically absent colons, those that were unable to complete the BBPS due to reasons other than bowel prep, and those with missing data. These patients were eliminated because they could not be scored using the BBPS, which was used as the primary measure for the manuscript. While this could have had an impact on the final outcome, the overall sample size was large enough that the conclusions can still be seen as a generalized outcome for a broader population. In addition, MiraLAX-Gatorade was found to have superior BBPS scores and tolerability, the impact of this is not clear because the minimum score before which the adenoma detection rate decreases and interval colorectal carcinoma development increases has not been determined. Because this study was conducted at a single site, there were also concerns about generalizability. This was somewhat offset by the large sample size and diverse patient cohorts. With regards to the assessment of bowel cleansing efficacy, since BBPS scores are assigned by the nurse who participated in the procedure, variability in scoring exists. This was addressed through the in-service training that was given to the nurses participating. The lack blinded reviewers of the BPPS scores could also be a flaw, but this made the study better reflect clinical practice and eliminate observer bias. Providers selecting regimens may have been biased but having 75 participating helps lessen this. Regression analysis with random effects accounted for differences in the providers such as endoscopic skills and experience, personal bowel cleansing thresholds, and patient population. Finally, the study was unable to account for patients who missed or canceled their procedure. Patients with intolerable preparations may have been part of this group, meaning tolerability and bowel cleansing effects of GoLYTELY may have been overestimated. There were no significant conflicts of interest, although the authors received a consulting fee from Bayer, which makes Miralax. This should not affect the results as providers chose the regimens.