Colonoscopy is used in people with chronic kidney disease receiving hemodialysis for renal replacement therapy and forms part of routine pre-transplant assessment.  Concerns exist about use of oral bowel-cleansing agents for colonoscopy in people receiving haemodialysis, despite guidelines recommendations.  It is plausible that intravascular depletion and consequent hypotension may cause thrombosis of the vascular access use for hemodialyisis.  We evaluated the comparative efficacy, safety and tolerability a low-volume versus a high-volume polyethylene glycol based bowel preparation for colonoscopy in people on hemodialysis. 

This was a multicentre, PROBE design,  parallel-arm, centrally randomized, non-inferiority phase IV trial.Eligible subjects were allocated 1:1 to either low-volume (2 liters) or high-volume (4 liters) bowel preparation. The primary outcome was the proportion of participants with adequate bowel cleansing, evaluated using the Boston Bowel Preparation Scale (BBPS). Secondary outcomes were adenoma detection rate (ADR), caecal intubation rate, patient tolerability/compliance, and adverse events (both related to colonoscopy and to dialysis/renal function). Differences were tested via t-test or  Mann–Whitney U test for continuous normal variables or through chi-square or Fisher exact test for categorical ones.

Sixty-nine patients were randomized in four centres, 35 to low volume and 34 to high volume groups.Mean age was 62.5±10.7 years; 54 were men and 15 were women. A total of 33 in the low-volume and 27 in high-volume completed colonoscopy; the main reasons for exclusion was stopping to take preparation (n=5).15.No difference in caecal intubation rate was observed (100% vs 96.6%, p=1). Adequate bowel preparation was achieved in 69.7% and 63.0% of low and high-volume groups, respectively (p=0.58). Mean BBPS was similar in the two groups (5.6±2.4vs 5.4±2.4, p=0.51). When sub-analyzing according to each bowel segment, we did not observe any difference for right, transverse nor left colon.

The ADR was 35.0% in low-volume and 37.0% in high volume (p=0.10).Median number of polyps per patient was 1.5 in low and 2 in high-volume group (p=0.45).

Mild abdominal swelling was the most common adverse event (27.3% vs 25.9% for high and low volume respectively, p=0.29), followed by mild nauseas (27.3% vs18.5%, p=0.29). A similar rate of willingness to repeat the same preparation was observed (87.9% vs 85.2%, p=0.76).No emergency dialysis sessions were required and no deaths were recorded. Levels of serum creatinine remained stable before and after preparation in both groups.

This small scale study based on an opportunistic sample confirms the difficulties in adequate preparation for colonoscopy in people on hemodialysis.  It shows that both a low and a high-volume preparation were equally effective and safe.We did not observea  better tolerability for low-volume preparations, which may be a key findings but also due to suboptimal small sample size.  Further large scale studies may be required to confirm our findings.