Melanosis coli: A naturally enhanced contrast for colonoscopy?

Melanosis coli, a dark brown discoloration of the colonic mucosa, is associated with chronic use of anthraquinone laxatives (senna). In addition, Aloe Vera¹ and Rhubarb,² used in traditional alternative medicine in Taiwan, are also implicated as the causes of melanosis coli. The term “melanosis,” however, may be misleading, since the pigment responsible for the discoloration is not the deposition of melanin but of lipofuscin in macrophages in the colonic lamina propria. Anthroquinone laxatives induce apoptosis of colon epithelium cells, which are ingested by adjacent macrophages within the mucosa. Those macrophages migrate to the lamina propria and convert the apoptotic cells into lipofuscin with lysozyme. Interestingly, adenomas have impaired absorption of apoptotic debris into the macrophages³ and thus stand out in the dark mucosa of the melanosis coli as a “pigmentation sparring sign.”

Whether melanosis coli increases the risk of colon adenoma and adenocarcinoma is controversial. An early prospective case-control study by Siegers et al suggested anthraquinone laxative abuse was associated with a relative risk of 3.04 (95% confidence interval [CI]: 1.18–4.90) for colorectal cancer.⁴ Most recent case-control studies, however, showed that melanosis coli was associated with increased detection of adenomas but not adenocarcinomas. Kassim et al found that patients with melanosis coli were more likely to have both hyperplastic polyp and low-grade adenoma, but not adenocarcinoma.⁵ Liu et al found that melanosis coli was associated with higher detection rates of low-grade adenoma (odds ratio [OR] = 1.54; 95% CI: 1.06–2.23; P < .05) but similar detection of high-grade adenomas or adenocarcinomas.⁶ Blackett et al showed patients with melanosis were more likely to have an adenoma ≤5 mm (OR = 1.62; 95% CI: 1.04–2.51; P = .03) but not adenomas 6 to 9 mm or ≥10 mm.⁷ Abu Baker et al even reported that melanosis coli was associated with less diagnosis of adenocarcinoma than controls (0.3% vs 3.9%; P < .001).⁸ Katsumata et al conducted a case-control study in Japan and performed a meta-analysis of five studies at the same time. They concluded that although hyperplastic polyps and adenomas were more frequently detected in patients with melanosis coli, the risk of colorectal cancer was not increased.⁹

There are two plausible explanations for the association between melanosis coli and increased detection of adenomas: either melanosis coli promotes the development of adenoma or it facilitates the detection of adenoma. The lack of association between melanosis coli and adenocarcinoma lends support to the latter assumption, as, in accordance with the theory of the colorectal adenoma-carcinoma sequence, increased development of adenoma is supposed to increase the risk of adenocarcinoma.

Going a step further than enhancing detection of adenomas, Chan et al proposed that the pigmentation sparring sign may help differentiate adenomas from hyperplastic polyps in this issue of Adv Dig Med.¹⁰ The authors conducted a retrospective review of 571 consecutive colonoscopies for cancer screening in 2014. These store endoscopic images (3–5 frames for each polyp) from patients with melanosis coli and polyps were reviewed by three experienced endoscopists with no prior access to the images. The pigmentation sparring sign was considered positive if two or more endoscopists were in agreement. Twenty-three patients (4%) had melanosis coli. Sixteen of these patients had 35 polyps (21 adenomas and 14 hyperplastic polyps). The average diameter of the adenomas was 3.9 mm (range, 2–15 mm). The use of the pigmentation sparing sign for predicting adenoma had a sensitivity of 95.2%, a specificity of 78.6%, a positive predictive value of 87.0%, and a negative predictive value (NPV) of 91.7%. The overall accuracy for predicting melanosis coli with the pigmentation sparing sign was 88.6%.

Although the reported accuracy of the pigmentation sparring sign is considerably lower than the modern state-of-the-art image enhancing stools, such as narrow band imaging without magnification (sensitivity and accuracy of 96% and 93%, respectively), the numbers are higher than the sensitivity and accuracy of high-definition white light without using any image enhancing technique (38% and 61%, respectively).¹¹ The results suggested that the pigmentation sparring sign might be a helpful tool for differentiating adenoma from hyperplastic polyp. It is also interesting to note that the pigmentation sparring sign meets the benchmark of the Preservation and Incorporation of Valuable Endoscopic Innovations (PIVI) criteria proposed in 2011 by the American Society for Gastrointestinal Endoscopy,¹² which require an NPV of >90% for diminutive hyperplastic rectosigmoid polyps to implement the leave-in-situ strategy. That is, if the results are confirmed by future studies, endoscopists can leave diminutive polyps alone in the rectosigmoid colon in patients with melanosis when the pigmentation sparring sign is absent.

Despite the potential significance of the findings, the results of the study must be interpreted with caution. First, out of 571 patients evaluated, only 23 patients (4%) had a diagnosis of melanosis coli. The prevalence was similar to those in the literature and suggested that only a minority of patients undergoing colonoscopies can benefit from the utilization of the sign. Second, the sample size of the current study was small, which reduces the power of the study and increases the margin of error. Third, the presence of the pigmentation sparring was determined by three endoscopists of unknown experience, and 4 of the 35 polyps were diagnosed by majority (two of the three endoscopists) instead of unanimity, suggesting the sign might be equivocal in some cases. Future prospective studies with a larger sample size involving endoscopists of various experience levels are warranted to verify the findings of the current study.

Most pathological studies reported on the lack of melanosis coli in colon adenoma,^{3, 13} Coyne et al, however, showed that melanosis coli could be seen in adenomatous polyps.¹⁴ To complicate the matters, some hyperplastic polyps also showed the pigmentation sparring sign.¹⁴ Recent case-control studies also suggest that hyperplastic polyps were more frequently detected in patients with melanosis coli in addition to adenomas,^{5, 9} suggesting the natural contrast-enhancing effect of the pigmentation sparring sign might also be present in hyperplastic polyps.

In conclusion, the pigmentation sparring sign might help endoscopists differentiate adenomatous polyps from hyperplastic polyps in addition to find more adenomas in patients with melanosis coli. Although the authors raised an intriguing question, the definite answer warrants further studies because of limitations of the study and conflicting existing literatures.

The author declars no conflict of interest.