You are here

  • Home
  • Archive
  • Volume 66, Issue 2
  • Split-dose preparation for colonoscopy increases adenoma detection rate: a randomised controlled trial in an organised screening programme

Article Text

Article menu

Download PDFPDF

Endoscopy
Original article
Split-dose preparation for colonoscopy increases adenoma detection rate: a randomised controlled trial in an organised screening programme
  1. F Radaelli1,
  2. S Paggi1,
  3. C Hassan2,
  4. C Senore3,
  5. R Fasoli4,
  6. A Anderloni5,
  7. F Buffoli6,
  8. M F Savarese6,
  9. G Spinzi1,
  10. D K Rex7,
  11. A Repici5
  1. 1Division of Digestive Endoscopy and Gastroenterology, Valduce Hospital, Como, Italy
  2. 2Endoscopy Unit, Nuovo Regina Margherita Hospital, Rome, Italy
  3. 3Centro di Prevenzione Oncologica (CPO Piemonte), AOU Città della Salute e della Scienza, Turin, Italy
  4. 4Division of Digestive Endoscopy, Imperia Hospital, Imperia, Italy
  5. 5Endoscopy Unit, IRCCS Istituto Clinico Humanitas, Rozzano, Milano, Italy
  6. 6Division of Digestive Endoscopy and Gastroenterology, A.O. Istituti Ospitalieri di Cremona, Cremona, Italy
  7. 7Division of Gastroenterology/Hepatology, Indiana University Medical Center, Indianapolis, Indiana, USA
  1. Correspondence to Dr F Radaelli, Division of Digestive Endoscopy and Gastroenterology, Valduce Hospital, Via Dante 11, Como 22100, Italy; francoradaelli01{at}gmail.com

Abstract

Objective Although a split regimen of bowel preparation has been associated with higher levels of bowel cleansing, it is still uncertain whether it has a favourable effect on the adenoma detection rate (ADR). The present study was aimed at evaluating whether a split regimen was superior to the traditional ‘full-dose, day-before’ regimen in terms of ADR.

Design In a multicentre, randomised, endoscopist-blinded study, 50–69-year-old subjects undergoing first colonoscopy after positive-faecal immunochemical test within an organised colorectal cancer organised screening programmes were 1:1 randomised to receive low-volume 2-L polyethylene glycol (PEG)-ascorbate solution in a ‘split-dose’ (Split-Dose Group, SDG) or ‘day-before’ regimen (Day-Before Group, DBG). The primary endpoint was the proportion of subjects with at least one adenoma. Secondary endpoints were the detection rates of advanced adenomas and serrated lesions at per-patient analysis and the total number of lesions.

Results 690 subjects were included in the study. At per-patient analysis, the proportion of subjects with at least one adenoma was significantly higher in the SDG than in the DBG (183/345, 53.0% vs 141/345, 40.9%, relative risk (RR) 1.22, 95% CI 1.03 to 1.46); corresponding figures for advanced adenomas were 26.4% (91/345) versus 20.0% (69/345, RR 1.35, 95% CI 1.06 to 1.73). At per-polyp analysis, the total numbers of both adenomas and advanced adenomas per subject were significantly higher in the SDG (1.15 vs 0.8, p <0.001; 0.36 vs 0.22, p<0.001).

Conclusions In an organised screening setting, the adoption of a split regimen resulted into a higher detection rate of clinically relevant neoplastic lesions, thus improving the effectiveness of colonoscopy. Based on such evidence, the adoption of a split regimen for colonoscopy should be strongly recommended.

Clinical trial registration number NCT02178033.

  • COLONOSCOPY
  • COLORECTAL ADENOMAS
  • COLORECTAL CANCER
  • COLORECTAL CANCER SCREENING
  • ENDOSCOPY

https://doi.org/10.1136/gutjnl-2015-310685

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Significance of this study

    What is already known on this subject?

    • A split regimen of bowel preparation for colonoscopy has been associated with an improvement of colonic cleansing.

    • A suboptimal adenoma detection rate (ADR) has been associated with a reduced efficacy of colonoscopy in preventing colorectal cancer (CRC) incidence and mortality.

    • The impact of a split regimen over the ADR of colonoscopy is uncertain.

    • Organised programmes with immunochemical faecal test represent an efficient and effective way to reduce CRC burden in the general population.

    What are the new findings?

    • In a randomised trial, a split regimen of bowel preparation increased the proportion of subjects with at least one adenoma and one advanced adenoma by 22% and 35%, respectively, when compared with a day-before regimen in the setting of an organised screening programme.

    • The adoption of a split regimen also resulted in higher numbers of both adenomas and advanced adenomas per subject at per-polyp analysis.

    • The level of colon cleansing was higher in the split compared with the day-before regimen.

    How might it impact on clinical practice in the foreseeable future?

    • The adoption of a split regimen for screening and diagnostic colonoscopy must be primarily recommended because of its efficacy in increasing the ADR when considering the strict association between the ADR and the reduction in CRC incidence and mortality at colonoscopy.

    Introduction

    Colorectal cancer (CRC) is a major cause of morbidity and mortality.1 ,2 CRC screening with biannual faecal immunochemical test (FIT) has been shown to reduce CRC mortality and incidence,3 ,4 and population-based organised screening programmes based on FIT have already been implemented in Italy and the Netherlands.5 ,6

    A suboptimal adenoma detection rate (ADR) has been strictly related to the risk of postcolonoscopy interval cancer and its mortality.7–13 Inadequate bowel preparation has been associated with a suboptimal ADR and higher risk for missing neoplastic lesions.14 ,15 In 75 569 FIT-positive subjects, an inadequate bowel preparation has been shown to reduce ADR by 35%.16 Inadequate bowel preparation has also been associated with higher risk of incomplete colonoscopy, as well as with early repetition of colonoscopy and surveillance anticipation, which in turn results in increased costs and reduced efficiency of screening colonoscopy.17–19

    Split regimens of bowel preparation have been associated with a higher level of colon cleansing.20 ,21 For this reason, they have been recommended by European and American Guidelines.22 ,23 However, there is still uncertainty on whether the higher level of cleansing associated with a split regimen also results in a higher ADR, which represents by far a more relevant quality indicator than the level of cleansing itself. Although a higher ADR with a split regimen was observed in retrospective series, biases related with the study design cannot be excluded.24 ,25

    This randomised investigator-blinded trial was designed to evaluate whether a ‘split regimen’ was superior to the traditional ‘full-dose, day-before regimen’ in terms of ADR in a FIT-based organised screening programme.

    Methods

    This investigator-blinded, randomised controlled study was conducted in four endoscopy centres in Northern Italy participating in the organised mass CRC screening programme. Written and informed consent was obtained from all subjects enrolled in the study. This was a no-profit study and the investigators had no relationship with the manufacturers of the bowel cleansing agent used in the study; no funding for the study was solicited or accepted. The trial was registered in the http://www.clinicaltrials.gov register (NCT02178033).

    CRC organised screening programme in Italy

    The organised screening programme with FIT in Italy is performed at a regional level. Screening centres in each region are responsible for inviting eligible subjects. Most regional programmes invite people aged 50–69 years to perform FIT on a biennial basis. Individuals with a positive FIT result (cut-off=100 ng Hb/mL buffer) are offered colonoscopy.

    Study population

    The target population included 690 subjects participating in the screening programme undergoing their first colonoscopy following a positive FIT from February 2014 to April 2015. We excluded from the study: (i) subjects not eligible for invitation in the screening programme (colonoscopy already performed in the previous 5 years, personal history of CRC, colonic adenomas or IBD, severe comorbidity, including end-stage cardiovascular, pulmonary, liver or renal disease); (ii) patients with previous colonic resection; (iii) patients on antithrombotic therapy, precluding polyp resection; (iv) patients at risk for inhalation; (v) patients with known hypersensitivity or contraindications to the study product (ie, phenylketonuria or glucose-6-phosphate dehydrogenase deficiency) and (vi) patients who were not able or refused to give informed written consent.

    Bowel cleansing solution and dietary restrictions

    All participants received low-volume (2 L) hypertonic polyethylene glycol (PEG) plus ascorbate solution (MOVIPREP*, Norgine, Harefield, UK), each litre containing 100.0 g macrogol 3350, 7.5 g sodium sulfate, 2.7 g sodium chloride, 1.0 g potassium chloride, 4.7 g ascorbic acid, 5.9 g sodium ascorbate and lemon or orange flavouring. Participants also received a standardised low-fibre diet before the colonoscopy, avoiding fruit, legumes and vegetables for 3 days before the procedure. They were allowed to have a standard breakfast and a light lunch on the day before the procedure, but no solid food was permitted since then. Clear liquids were permitted until 2 h before the procedure.

    Randomisation and blinding

    Participants were randomised in a 1:1 ratio to receive low-volume 2 L PEG plus ascorbate solution in a ‘split-dose’ (study arm) or in a ‘day-before’ regimen (control arm) based on a computer-generated random sequence. A randomisation list for each participating centre was produced by the coordinating centre via computer-generated treatment code list using SAS (statistical software package V.9.1); sealed envelopes containing the indication of the randomisation arm were delivered to each centre. Unblinded staff members (nurses or physicians), who were not subsequently involved in the endoscopy procedures, accomplished treatment allocation during the precolonoscopy screening visit and provided oral and written information on diet and cleansing solution intake. Data on patient compliance, tolerability and acceptability were collected on the day of colonoscopy, immediately before the procedure, by a nurse through a standardised questionnaire. To ensure masking, the investigators who performed colonoscopy were involved neither in the randomisation process nor in the preprocedure data collection. Besides, both patients and nurses were recommended not to disclose the type of bowel cleansing regimen to the investigator.

    Subjects randomised to ‘split-dose’ regimen (Split-Dose Group, SDG) were instructed to take the first litre (one MOVIPREP sachet dissolved in 1 L of water, according to manufacturer's instruction) on the evening before colonoscopy at about 20:00 and the second one on the day of the procedure about 4 h before the scheduled procedure time.

    Those randomised to ‘day-before’ regimen (Day-Before Group, DBG) were instructed to take the entire solution the evening before colonoscopy, the first litre between 18:00 and 19:00 and the second one between 21:00 and 22:00.

    Subjects in both groups were also encouraged to drink at least 0.5 L of clear liquids after each dose of cleansing solution, according to manufacturer's indications.

    Data collection and colonoscopy procedure

    On the day of colonoscopy, before starting the procedure, patients’ compliance (optimal=intake of the whole solution; fair=intake ≥75% of the solution; poor=intake of <75% of the solution) was assessed by an unblinded nurse by means of a structured questionnaire. Patients were also asked to rate the difficulty in the solution intake, the quality of sleep the night before colonoscopy and intensity of side effects (nausea or vomiting, bloating, abdominal pain) by means of a 4-point semiquantitative scale (3=severe distress; 2=moderate distress; 1=mild distress; 0=no distress at all). Patients were also asked about the need of travel interruption and the occurrence of faecal incontinence during the transfer to the endoscopy service and whether they would be willing to repeat the same preparation regimen in future (yes or no). Colonoscopies were performed with adult colonoscopes, by experienced endoscopists (two investigators per centre), with credentialing for participating in the CRC screening programmes (more than 5000 colonoscopies overall and/or more than 300 colonoscopies in the previous year). Sedation was managed according to the practice protocol of each centre and was selected for each procedure, according to the preferences of the endoscopist and of the subject; colonoscopy could be performed without sedation, when requested or required (ie, the absence of an escort), under conscious (intravenous midazolam±meperidine or fentanyl) or deep sedation (propofol). In order not to introduce any bias, the systematic use of antispasmodics was not allowed in study participants; any occasional use to facilitate caecal intubation was left at the discretion of the endoscopist and had to be recorded in patient form.

    All colonoscopies, regardless of the randomisation arm, were scheduled in the second part of the morning (from 10:00 to 12:00), in order to maximise the compliance with split-dose regimen and not to penalise the ‘day-before’ regimen by an excessively long runway time. Caecal intubation was confirmed by documentation of caecal landmarks (caecal valve and appendix orifice or intubation of terminal ileum). During withdrawal of the colonoscope, the colonic mucosa was carefully inspected. The withdrawal time for diagnostic procedures was targeted to a minimum of 6 min in accordance with colonoscopy quality practice, and it was recorded for each procedure.

    All detected polyps were directly removed and obtained for histological assessment, with the possible exception of diminutive (less than 5 mm) rectal polyps with typical appearance of hyperplastic lesions. Such lesions were treated according to the endoscopists’ standard practice, and in case of multiple, diminutive polyps in the rectum, at least one was removed for the purposes of pathological confirmation. The morphology of polyps was classified according to Paris classification.26 Lesions 0-II, according to Paris classification, were considered as flat. Polyp size was estimated by the endoscopist by comparing it with the span of an open biopsy forceps, the sheath of a polypectomy snare or the diameter of an open snare placed against the polyp. However, pathologist's measure, when available, was considered the reference standard.27 Polyps were considered proximal whenever they were located proximally to the splenic flexure (caecum, ascending colon and transverse colon).

    The quality of bowel cleansing was evaluated and graded during the procedure after washing and suctioning by the blinded endoscopist performing the exam through the validated Harefield Cleansing Scale (HCS).28 Briefly, 5-point qualitative scale (0=irremovable, heavy, hard stool; 4=empty and clean) is applied individually to five separate colon segments (rectum, sigmoid, descending colon, transverse colon and ascending colon). The results can be aggregated into four grades (A–D) or condensed into a binary assessment of cleansing (successful/unsuccessful). According to the scale, a successful cleansing was considered in case of grade A (all segments scoring at least 3) or grade B preparation (at least one segment scoring 2) (figure 1).

    Figure 1
    Figure 1

    The Harefield Cleansing Scale.

    Histopathology

    All resected lesions, either by forceps or snare, were sent to pathology in separated jars and were processed and stained for histopathology using standard methods and evaluated by expert pathologists (one in each centre) according to the Vienna criteria.29 All lesions were classified as inflammatory/normal mucosa, or adenoma, or serrated (hyperplastic, sessile serrated polyp (SSP), traditional serrated adenoma (TSA)). An advanced adenoma was defined as an adenoma ≥10 mm and/or with villous component of at least 20% and/or high-grade dysplasia.

    Outcome measures

    The primary outcome measure was the ADR, defined as the proportion of participants with at least one adenoma (per-patient analysis).

    Secondary outcomes were:

    1. The advanced ADR, defined as the proportion of participants with at least one advanced adenoma at a per-patient analysis.

    2. The number of adenomas, advanced adenomas and SSP per subject, defined as the total number of detected lesions in each group divided by the total number of participants (per-polyp analysis).

    3. The number of proximal and flat adenomas, defined as the total number of detected lesions in each group divided by the total number of participants (per-polyp analysis).

    4. The rate of overall and proximal colon successful cleansing, as assessed by the endoscopists by the HCS

    5. Patient compliance, acceptability and tolerability as assessed by a nurse before colonoscopy.

    Sample size calculation and statistical analysis

    The expected prevalence of adenomas in the DBG was 35%, based on data from the national screening programme. We aimed to detect a 10% absolute increase in adenoma detection in the SDG, resulting in an expected ADR of 45%. With a two-sided significance level of 0.05 and a power of 0.80, 345 subjects per arm had to be included in the study. Taking into account an expected 5% drop-out rate, we planned to enrol 725 subjects; study enrolment was stopped after achieving the target sample in each arm.

    Patients who were randomised but did not attend their colonoscopy appointment or did not take the study product for bowel cleansing were not included in the primary analysis.

    Categorical variables were summarised using frequencies and percentages, while quantitative variables were summarised using means and SDs. Fisher's exact test and χ2 test for trend were used to compare categorical variables, whereas Student's t test was used for continuous variables. Analyses were carried out with the SAS statistical software package V.9.1 using log-binomial regression for multivariable estimation of prevalence ratios. Differences were expressed as relative risk (RR) with 95% CIs. Adjustments were made in the multivariable analysis for age, gender, colonoscopy completion, screening history (first vs subsequent tests), quality of bowel preparation, type of instrument (high definition vs standard) and screening centre.

    All statistical tests were two-sided and were considered statistically significant at p<0.05.

    The results were reported according to CONsolidated Standards of Reporting Trials (CONSORT) guidelines.30

    Results

    In the study period, a total of 833 subjects were evaluated; of them, 711 were considered eligible and were 1:1 randomised to SDG or DBG; overall, 21 subjects dropped out after randomisation (14 missed the colonoscopy appointment, 7 did not follow prescriptions and took an alternative product for bowel cleansing), so that 345 subjects per arm were finally included in the study. Subject flow is represented in figure 2. Groups were comparable with respect to age, gender, screening history (first vs subsequent FIT rounds) and risk factors for difficult bowel preparation (ie, chronic constipation with the use of laxatives). In five cases (two in the SDG and three in the DBG), the endoscopist was unintentionally informed (unblinding) by patient or nurse on bowel prep regimen before or during the procedure (table 1).

    View this table:
    Table 1

    Baseline demographics and colonoscopy data

    Figure 2
    Figure 2

    Study flow chart.

    Colonoscopy procedure and findings

    Caecal intubation was achieved in 335 subjects (97.1%) in the SDG versus 329 (95.4%) in the DBG (p=0.32) (table 1). The two groups were well balanced with respect to sedation practice (conscious sedation in 93.3% in the SDG vs 93.0% in the DBG, p=0.45) and to withdrawal time (mean 8.5, SD 1.9 min in the SDG and mean 8.7, SD 2.0 min in the DBG, p=0.43). A total of 12 adverse events occurred (eight postpolypectomy bleedings and four cardiorespiratory) and no difference was observed between the two study arms. All bleeding and cardiorespiratory events were managed successfully during the procedure and were uneventful.

    As concerns colonoscopy findings, a total of 907 polypoid or non-polypoid lesions were detected; of them, 674 were adenomas, 161 hyperplastic and 48 serrated (45 SSP, 3 TSA); the remaining 24 lesions were inflammatory. Among adenomas, 201 (29.8%) had advanced features, 302 (44.8%) were located in the proximal colon and 56 (8.3%) were flat. A total of 42 invasive carcinomas were found; of them, 23 were in the SDG and 19 in the DBG, respectively.

    Adenoma detection

    Overall, 324 (46.9%) and 160 (23.1%) patients had at least one adenoma and one advanced adenoma, respectively. The distribution of adenomas across SDG and DBG according to histology, localisation and morphology is shown in table 2.

    View this table:
    Table 2

    Adenoma prevalence and characteristics by study arm

    At a per-patient analysis, the proportion of subjects with at least one adenoma (ADR) was significantly higher in the SDG than in the DBG (53.0% vs 40.9%, RR 1.22, 95% CI 1.03 to 1.46); corresponding figures for advanced adenomas were 26.4% versus 20.0% (RR 1.35, 95% CI 1.06 to 1.73) (figure 3).

    Figure 3
    Figure 3

    Proportion of subjects with at least one adenoma, advanced adenoma and sessile serrated polyps (per-patient analysis). SSP, sessile serrated polyp.

    At a per-polyp analysis (table 2), the total numbers of both adenomas and advanced adenomas per subject were significantly higher in the SDG than in the DBG. The same figures were also observed for right-sided and flat adenomas.

    Sessile serrated lesion detection

    A total of 45 SSPs were observed in 41 subjects; of them, 32 SPPs were detected in SDG. There was a trend towards significance in favour of SDG as concerns either the number of patients with at least one SSP (7.8% vs 4.1%, p=0.053) or the total number of SSPs per subject (0.09±0.47 vs 0.04±0.19, p=0.06).

    Surveillance interval recommendations

    Based on adenoma and serrated lesion detection, we calculated the surveillance colonoscopy intervals in the two groups. According to current guidelines by the European Society of Gastrointestinal Endoscopy (ESGE), the number of high-risk subjects (those with ≥3 adenomas, 1 adenoma ≥10 mm, any adenoma with advanced histology, any serrated lesions ≥10 mm or with dysplasia) who were recommended a 3-year surveillance interval were 122 and 91 in the SDG and DBG, respectively (35.4% vs 26.4%, p=0.013).

    Quality of bowel preparation

    The quality of bowel cleansing for either the whole colon or the right colon in each study arm is detailed in table 3. The proportion of patients with successful colon cleansing (HCS grade A or B) was significantly higher in the SDG than in the DBG, either when considering the whole (95.4% vs 89.0%, p=0.001) or the right colon (94.5% vs 88.1%, p=0.002). Consistent results were observed when considering the proportions of very good/excellent preparations (grade A) for either the whole (79.7% vs 54.8%, p<0.001) or the proximal colon (80.2% vs 54.7%, p<0.001).

    View this table:
    Table 3

    Quality of bowel cleansing by study arm

    Compliance and tolerability

    Significantly better compliance and tolerability endpoints were observed in the SDG (tables 4 and 5). Overall, 19 (2.8%) and 5 (0.7%) patients reported the need of travel interruption and the occurrence of faecal incontinence during the transfer to the endoscopy service; 603 (87.4%) reported they would be willing to repeat the same preparation regimen in future. No difference was observed in the SDG and DBG for these figures.

    View this table:
    Table 4

    Compliance, tolerance and acceptance by study arm

    View this table:
    Table 5

    Adverse events

    Discussion

    In a randomised trial carried out in an organised CRC screening programme, we showed that the adoption of a split regimen of bowel preparation was significantly and independently associated with a substantial increase of the ADR at per-patient analysis. In addition, we showed that a split regimen was also associated with a significant increase in the detection of advanced neoplasia at per-patient analysis, as well as with an increase in the mean number of adenomas, advanced adenomas, proximal and flat lesions, including SSP, at per-polyp analysis. The results of our study are relevant for the following reasons.

    First, we showed that the implementation of a split regimen should be primarily recommended due to its favourable effect on ADR rather than on the level of cleansing, as currently done in the available guidelines. A low ADR has been related with a higher risk of interval CRC and associated mortality,10 ,13 ,31 while no conclusive evidence has been found with the level of bowel preparation.32 In addition, ADR represents a very solid endpoint due to the very low interpathology variability in the differential diagnosis between neoplastic and non-neoplastic lesions,33 while the assessment of the level of cleansing is hampered by unavoidable degree of subjectivity and higher degree of interoperator variability. Thus, our data directly associated the split regimen with the main proxy of the efficacy of colonoscopy in reducing CRC incidence and mortality, offering a different and superior rationale to its recommendation in clinical practice. The randomised setting coupled with the highly controlled and homogeneous setting of an organised FIT programme also brought to a much higher level of evidence the initial findings on such association shown in retrospective series.24 ,25

    Second, we showed that the effect of a split regimen on ADR also resulted in a higher detection of advanced neoplasia at per-patient analysis. Advanced neoplasia represents the main target of FIT-based organised screening programmes, due to the very high prevalence of these lesions in FIT-positive subjects and the high risk of progression of advanced adenomas in invasive cancer.5 ,34 ,35 Thus, the adoption of a split regimen primarily and directly increases the efficacy of population-based FIT programmes by reducing the rate of false-negative results due to inadequate bowel preparation.

    Third, the higher rate of advanced neoplasia and clinically relevant serrated lesions coupled with the increase of mean lesions per patient—that is, a higher rate of multiplicity—resulted into a more intensive postpolypectomy surveillance schedule compared with the ‘day-before’ arm. Although the additional benefit of surveillance for patients with multiple small (<10 mm) low-risk adenomas (making up about 30% of the additional referrals in the SDG) is still uncertain, a substantial reduction of CRC incidence compared with the efficacy of the index polypectomy has been documented among high-risk patients undergoing postpolypectomy surveillance.36 ,37 Thus, our data put in direct connection the implementation of a split regimen with the adoption of a clinically relevant surveillance strategy.

    Fourth, we showed that the favourable effect of a split regimen on the detection of neoplastic lesions was associated with a better cleansing level compared with the ‘day-before’ arm, providing a biological plausibility to the efficacy of a split regimen on ADR. We may postulate that the improved cleansing led to a more meticulous exploration of the colorectal mucosa, increasing the detection of polypoid and flat lesions at both per-patient and per-polyp analyses. Our data are in line with the recent results of two meta-analyses showing a higher cleansing level when adopting a split regimen.20 ,21

    Fifth, a ‘split’ regimen appeared to be more tolerable than a ‘day-before’ regimen, potentially increasing the long-term acceptability of colonoscopy within an organised screening programme. Moreover, we found a very rare occurrence of travel interruptions and faecal incontinence episodes; these rates were higher in the split than in the full-dose day-before regimen, although this difference was not statistically significant. However, the inclusion in the study of subjects scheduled for late morning procedures only (10:00—12:00) may have emphasised the acceptance and tolerability of split-dose preparation, as it is known that they may be lower for early morning examinations.

    There are limitations to the present analysis. The observed prevalence of adenomas in both arms was slightly higher than that expected in the sample size calculation, possibly due to the controlled setting of the trial compared with real life. However, such minor difference did not affect the validity of our analysis. The inclusion of a high-risk population may represent a limitation to the generalisability of the study to average-risk patients; further studies are needed in those patients. Similarly, we used a low-volume PEG regimen in both arms, so that it may be uncertain whether these results can be extrapolated to high-volume PEG or non-PEG regimens. However, at least from data on cleansing level, a split regimen appears to be effective, regardless of the adopted laxative.20 The choice of the Harefield Cleansing Scale for our secondary endpoint—that is, the quality of cleansing—may be questionable, as it does not have a widespread use in clinical practice. However, it is a validated scale,28 and it is the only one, together with the Boston Bowel Preparation Scale, which evaluates bowel cleansing after washing and suctioning. Finally, we cannot exclude the possibility that blinding was impaired by the preparation quality itself, since split-dosing reduces the amount of adherent mucus in the proximal colon. Colonoscopists could have guessed which preparation was used by the appearance of the right colon, and we did not measure their estimates of the preparation used. However, withdrawal times for diagnostic procedures were comparable, showing that the examination effort was equivalent in the two study arms.

    In conclusion, in a high-quality multicentre randomised trial, we showed that the adoption of a low-volume split regimen resulted into a higher detection rate of clinically relevant neoplastic lesions, which are well known to be associated with a higher CRC protection rate. Such evidence should become the primary reason for recommending the adoption of a split regimen for colonoscopy.

    References

    1. ↵
      1. Edwards BK,
      2. Ward E,
      3. Kohler BA, et al
      . Annual report to the nation on the status of cancer, 1975–2006, featuring colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. Cancer 2010;116:544–73. doi:10.1002/cncr.24760
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Ferlay J,
      2. Parkin DM,
      3. Steliarova-Focher E
      . Estimates of the cancer incidence and mortality in Europe in 2008. Eur J Cancer 2010;46:765–81. doi:10.1016/j.ejca.2009.12.014
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Ventura L,
      2. Mantellini P,
      3. Grazzini G, et al
      . The impact of immunochemical faecal occult blood testing on colorectal cancer incidence. Dig Liv Dis 2014;46:82–6. doi:10.1016/j.dld.2013.07.017
      OpenUrlCrossRef
    4. ↵
      1. Zorzi M,
      2. Fedeli U,
      3. Schievano E, et al
      . Impact on colorectal cancer mortality of screening programmes based on the faecal immunochemical test. Gut 2015;64:784–90. doi:10.1136/gutjnl-2014-307508
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Zorzi M,
      2. Fedato C,
      3. Grazzini G, et al
      . Screening for colorectal cancer in Italy, 2010 survey. Epidemiol Prev 2012;36(6 Suppl 1):55–77.
      OpenUrlPubMed
    6. ↵
      1. Denters MJ,
      2. Deutekom M,
      3. Fockens P, et al
      . Implementation of population screening for colorectal cancer by repeated fecal occult blood test in the Netherlands. BMC Gastroenterol 2009;9:28. doi:10.1186/1471-230X-9-28
      OpenUrlCrossRefPubMed
    7. ↵
      1. Singh H,
      2. Nugent Z,
      3. Demers AA, et al
      . Rate and predictors of early/missed colorectal cancers after colonoscopy in Manitoba: a population-based study. Am J Gastroenterol 2010;105:2588–96. doi:10.1038/ajg.2010.390
      OpenUrlCrossRefPubMed
    8. ↵
      1. Singh H,
      2. Nugent Z,
      3. Demers AA, et al
      . The reduction in colorectal cancer mortality after colonoscopy varies by site of the cancer. Gastroenterology 2010;139:1128–37. doi:10.1053/j.gastro.2010.06.052
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Singh H,
      2. Turner D,
      3. Xue L, et al
      . Risk of developing colorectal cancer following a negative colonoscopy examination: evidence for a 10-year interval between colonoscopies. JAMA 2006;295:2366–73. doi:10.1001/jama.295.20.2366
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Baxter NN,
      2. Sutradhar R,
      3. Forbes SS, et al
      . Analysis of administrative data finds endoscopist quality measures associated with postcolonoscopy colorectal cancer. Gastroenterology 2011;140:65–72. doi:10.1053/j.gastro.2010.09.006
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Kaminski MF,
      2. Regula J,
      3. Kraszewska E, et al
      . Quality indicators for colonoscopy and the risk of interval cancer. N Engl J Med 2010;362:1795–803. doi:10.1056/NEJMoa0907667
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Rogal SS,
      2. Pinsky PF,
      3. Schoen RE
      . Relationship between detection of adenomas by flexible sigmoidoscopy and interval distal colorectal cancer. Clin Gastroenterol Hepatol 2013;11:73–8. doi:10.1016/j.cgh.2012.08.002
      OpenUrlCrossRefPubMed
    13. ↵
      1. Corley DA,
      2. Jensen CD,
      3. Marks AR, et al
      . Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 2014;370:1298–306. doi:10.1056/NEJMoa1309086
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Lebwohl B,
      2. Kastrinos F,
      3. Glick M, et al
      . The impact of suboptimal bowel preparation on adenoma miss rates and the factors associated with early repeat colonoscopy. Gastroint Endosc 2011;73:1207–14. doi:10.1016/j.gie.2011.01.051
      OpenUrlCrossRef
    15. ↵
      1. Chokshi RV,
      2. Hovis CE,
      3. Hollander T, et al
      . Prevalence of missed adenomas in patients with inadequate bowel preparation on screening colonoscopy. Gastroint Endosc 2012;75:1197–203. doi:10.1016/j.gie.2012.01.005
      OpenUrlCrossRef
    16. ↵
      1. Zorzi M,
      2. Senore C,
      3. Da Re F, et al
      . Quality of colonoscopy in an organised colorectal cancer screening programme with immunochemical faecal occult blood test: the EQuIPE study (evaluating quality indicators of the performance of endoscopy). Gut 2015;64:1389–96.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Hillyer GC,
      2. Basch CH,
      3. Lebwohl B, et al
      . Shortened surveillance intervals following suboptimal bowel preparation for colonoscopy: results of a national survey. Int J Colorectal Dis 2013;28:73–81. doi:10.1007/s00384-012-1559-7
      OpenUrlCrossRefPubMed
    18. ↵
      1. Rex DK,
      2. Imperiale TF,
      3. Latinovich DR, et al
      . Impact of bowel preparation on efficiency and cost of colonoscopy. Am J Gastroenterol 2002;97:1696–700. doi:10.1111/j.1572-0241.2002.05827.x
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Macphail ME,
      2. Hardacker KA,
      3. Tiwari A, et al
      . Intraprocedural cleansing work during colonoscopy and achievable rates of adequate preparation in an open-access endoscopy unit. Gastroint Endosc 2015;81:525–30. doi:10.1016/j.gie.2014.05.002
      OpenUrlCrossRef
    20. ↵
      1. Bucci C,
      2. Rotondano G,
      3. Hassan C, et al
      . Optimal bowel cleansing for colonoscopy: split the dose! A series of meta-analyses of controlled studies. Gastroint Endosc 2014;80:566–76. doi:10.1016/j.gie.2014.05.320
      OpenUrlCrossRef
    21. ↵
      1. Martel M,
      2. Barkun AN,
      3. Menard C, et al
      . Split-dose preparations are superior to day-before bowel cleansing regimens: a meta-analysis. Gastroenterology 2015;149:79–88. doi:10.1053/j.gastro.2015.04.004
      OpenUrlCrossRefPubMed
    22. ↵
      1. Hassan C,
      2. Bretthauer M,
      3. Kaminski MF, et al
      . Bowel preparation for colonoscopy: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 2013;45:142–50. doi:10.1055/s-0032-1326186
      OpenUrlCrossRefPubMed
    23. ↵
      1. Johnson DA,
      2. Barkun AN,
      3. Cohen LB, et al
      . Optimizing adequacy of bowel cleansing for colonoscopy: recommendations from the US multi-society task force on colorectal cancer. Gastroenterology 2014;147:903–24. doi:10.1053/j.gastro.2014.07.002
      OpenUrlCrossRefPubMed
    24. ↵
      1. Gurudu SR,
      2. Ramirez FC,
      3. Harrison ME, et al
      . Increased adenoma detection rate with system-wide implementation of a split-dose preparation for colonoscopy. Gastrointest Endosc 2012;76:603–8. doi:10.1016/j.gie.2012.04.456
      OpenUrlCrossRefPubMed
    25. ↵
      1. Jover R,
      2. Zapater P,
      3. Polania E, et al
      . Modifiable endoscopic factors that influence the adenoma detection rate in colorectal cancer screening colonoscopies. Gastroint Endosc 2013;77:381–9. doi:10.1016/j.gie.2012.09.027
      OpenUrlCrossRef
    26. ↵
      [No authors listed]. The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon: November 30 To December 1, 2002. Gastroint Endosc 2003;58:S3–43. doi:10.1016/S0016-5107(03)02159-X
      OpenUrlCrossRef
    27. ↵
      1. Von Karsa L,
      2. Patnick J,
      3. Segnan N, et al
      . European guidelines for quality assurance in colorectal cancer screening and diagnosis: overview and introduction to the full supplement publication. Endoscopy 2013;45:51–9. doi:10.1055/s-0032-1325997
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Halphen M,
      2. Heresbach D,
      3. Gruss HJ, et al
      . Validation of the Harefield Cleansing Scale: a tool for the evaluation of bowel cleansing quality in both research and clinical practice. Gastroint Endosc 2013;78:121–31. doi:10.1016/j.gie.2013.02.009
      OpenUrlCrossRef
    29. ↵
      1. Schlemper RJ,
      2. Riddell RH,
      3. Kato Y, et al
      . The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47:251–5. doi:10.1136/gut.47.2.251
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Boutron I,
      2. Moher D,
      3. Altman DG, et al
      . Extending the consort statement to randomized trials of nonpharmacologic treatment: explanation and elaboration. Ann Int Med 2008;148:295–309. doi:10.7326/0003-4819-148-4-200802190-00008
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Cooper GS,
      2. Xu F,
      3. Barnholtz Sloan JS, et al
      . Prevalence and predictors of interval colorectal cancers in medicare beneficiaries. Cancer 2012;118: 3044–52. doi:10.1002/cncr.26602
      OpenUrlCrossRefPubMed
    32. ↵
      1. Farrar WD,
      2. Sawhney MS,
      3. Nelson DB, et al
      . Colorectal cancers found after a complete colonoscopy. Clin Gastroenterol Hepatol 2006;4:1259–64. doi:10.1016/j.cgh.2006.07.012
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Rex DK,
      2. Alikhan M,
      3. Cummings O, et al
      . Accuracy of pathologic interpretation of colorectal polyps by general pathologists in community practice. Gastroint Endosc 1999;50:468–74. doi:10.1016/S0016-5107(99)70067-2
      OpenUrlCrossRef
    34. ↵
      1. Hol L,
      2. Kuipers EJ,
      3. Van Ballegooijen M, et al
      . Uptake of faecal immunochemical test screening among nonparticipants in a flexible sigmoidoscopy screening programme. Int J Cancer 2012;130:2096–102. doi:10.1002/ijc.26260
      OpenUrlCrossRefPubMed
    35. ↵
      1. Segnan N,
      2. Senore C,
      3. Andreoni B, et al
      . Comparing attendance and detection rate of colonoscopy with sigmoidoscopy and fit for colorectal cancer screening. Gastroenterology 2007;132:2304–12. doi:10.1053/j.gastro.2007.03.030
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. Cottet V,
      2. Jooste V,
      3. Fournel I, et al
      . Long-term risk of colorectal cancer after adenoma removal: a population-based cohort study. Gut 2012;61:1180–6. doi:10.1136/gutjnl-2011-300295
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Loberg M,
      2. Kalager M,
      3. Holme O, et al
      . Long-term colorectal-cancer mortality after adenoma removal. N Engl J Med 2014;371:799–807. doi:10.1056/NEJMoa1315870
      OpenUrlCrossRefPubMedWeb of Science
    View Abstract

    Footnotes

    • Contributors FR, SP, CS, CH, AR: study concept and design; interpretation of results; drafting of the manuscript. FR, SP, CS: acquisition of data and statistical analysis. All the authors: critical revision of the article for important intellectual content and final approval of the article.

    • Competing interests None declared.

    • Ethics approval The protocol was approved by the Valduce Hospital ethics committee and afterwards by all other participating institutions.

    • Provenance and peer review Not commissioned; externally peer reviewed.