StATS: Searching for information about the molasses with milk enema (October 15, 2005)

I am part of an informal group at Children's Mercy Hospital that is trying to examine the process of Evidence Based Medicine to try to better understand it ourselves so that we can teach it and incorporate it into our practice. About a month ago, a question came up:

In an otherwise healthy child with constipation what is the harm of giving a milk with molasses enema?

I was just about to head out on a vacation, but I did a quick Google search and came up with some surprising references. But now that I have a bit more time, I thought it would be good to research the question carefully.

As I had noted in earlier weblog entries, it is usually a good idea to search high level resources like BestBets and the AHRQ National Guideline Clearinghouse. Since I didn't expect to find anything at either resource, I thought I'd search on a single term, enema, to verify that nothing even remotely connected with the above question existed at these places.

The Best Bets web site returned 0 hits, but the Guideline Clearinghouse had 42 hits, most of them associated with colorectal cancer and endoscopy. But one guideline seemed just perfect: Constipation in infants and children: evaluation and treatment.

Unfortunately, the only discussion about enemas were the following two statements:

In infants, rectal disimpaction can be carried out with glycerin suppositories. Enemas are to be avoided [II-3].

In children, disimpaction may be carried out with either oral or rectal medication, including enemas [II-3].

The bracketed notations are the categories of the quality of evidence and the guideline described these categories as:

The level II-3 is a fairly weak form of evidence, just barely above expert opinions.

A similar guideline for adults was not at all helpful with regard to the milk with molasses enema, but did mention some serious issues with sodium phosphate enemas.

The Food and Drug Administration in 1998 also recommended package size restrictions and modifications in labeling of rectal enema sodium phosphate products because of reported serious side effects and reports of overdosing ("Laxative drug products," 1999). Labeling now must include warning statements regarding use with patients with a colostomy, congenital megacolon, imperforate anus, impaired renal function, heart disease, congestive heart failure, preexisting electrolyte disturbances, or in patients using diuretics that may affect electrolyte levels (Food and Drug Administration, HHS. Final Rule, 1999).

The frequent use of enemas, laxatives, and stool softeners is believed to lead to increased constipation in older adults

Chronic use of laxatives, stool softeners, and enemas by elderly persons has been associated with several significant clinical disorders, including diarrhea, hypermagnesemia, life-threatening hyperphosphatemia, hypoalbuminemia, an increased risk of fecal incontinence and perianal soiling, and poor response to bowel preparation for barium enema.

A third guideline, Functional constipation and soiling in children, also had a general warning:

Enemas have a risk of mechanical trauma. Saline and hypertonic phosphate may cause abdominal cramping. Hypertonic phosphate has risk of hyperphosphatemia, hypokalemia, and hypocalcemia, and may not be recommended for children under 4 years of age. If enemas are considered for infants, administer first in doctor’s office.

A search at Cochrane on the word "enema" yielded five hits, but none of them were relevant.

A search through PubMed yielded 8,106 publications when searching for enema, 827 when searching for molasses, and 69,842 when searching for milk. The intersection of these three terms yielded a single publication (and this was before I had a chance to apply a filter for children and infants):

Cardiopulmonary compromise associated with milk and molasses enema use in children. Walker M, Warner BW, Brilli RJ, Jacobs BR. J Pediatr Gastroenterol Nutr 2003: 36(1); 144-8. [Medline] [Abstract] [Full text] [PDF] OBJECTIVE: Constipation is a common childhood problem that frequently confronts pediatric practitioners. When enema treatment is indicated to treat constipation, milk and molasses enemas are sometimes used. Side effects and adverse outcomes have been described with many types of enemas, but not with milk and molasses. The purpose of this study is to describe five children who acutely decompensated after receiving milk and molasses enemas. METHODS: Retrospective review of the medical records of children with constipation who received milk and molasses enema with subsequent PICU admission at Cincinnati Children's Hospital Medical Center between 1994 and 2000. RESULTS: Five children developed significant hemodynamic deterioration after receiving milk and molasses enemas. One of these children died and the others recovered after aggressive resuscitation. CONCLUSIONS: Milk and molasses enema administration can be associated with significant hemodynamic compromise. These risks appear similar to those risk observed with the use of other enema preparations.

I normally don't reproduce the entire abstract on these web pages out of respect for the copyright of the individual journals, but here, the content is important to help you gauge the quality of evidence. This is a case series, which is a rather weak form of evidence. A particular striking problem is the difficulty estimating the NNH (Number Needed to Harm) or in contrasting the risk of this type of enema to other types. Reading through the article, I don't see how the authors justify the statement "These risks appear similar to those risk observed with the use of other enema preparations." At best, you can argue that the milk with molasses enema causes harm through similar mechanisms as other types of enemas, but perhaps milk with molasses is a milder and less risky type of enema or maybe it is far more toxic. The data in this paper does not allow you to discern the relative severity of risk.

Interestingly enough, when I went to Google Scholar to see if there was a version of this paper online, I found several other papers that were not included in the PubMed list. None seemed quite as current or as relevant as the citation listed above, but they were still interesting.

A regular Google search is also interesting, and I'll report on what that found if I get some free time.

Postscript (October 26, 2005): I re-reviewed the Google Scholar search and only one reference looked helpful. It is

This paper is a subjective review of the evidence because the authors did not provide any information on how they searched the literature for relevant information (strike one). It is also not peer-reviewed (strike two), and it appears on a commercial web site that promotes "Colon Cleansing/Hydration (also known as Colonics)." It is difficult to ignore conflict of interest for a subjective review of the evidence--would you trust a review of the Cox-2 inhibitors that was published on a web site sponsored by Merck? Strike three!

The Google search also yielded a lot of links that had obvious commercial biases. The first link, for example, that Google found:

was on a web site that has a links on its main page to

A commercial bias by itself is not bad. But a commercial bias associated with a subjective evaluation (rather than an objective study) is very troublesome. An objective study, such as a randomized trial or a systematic overview, is transparent and  you can gauge all of the inputs into the process. Equally important, an objective study is repeatable, which allows a disinterested party the possibility of replicating the research. Unfortunately, with a subjective review, you cannot gauge the quality of the inputs nor can you replicate the results unless you can find a way to crawl inside the heads of the original authors.

Even with commercial biases, a subjective review has some value. You can look up the references provided in the bibliography (though you cannot look up the references that the authors decided in their wisdom not to include). You can also discover issues that you may not have been aware of. For example, the Richards et al review listed four possible side effects of enemas (before dismissing them as trivial). Prior to reading this review, I had been aware of only two of those side effects.

Finally, any facts that work against the authors' commercial interests have some level of credibility. If the authors' grudgingly admit a problem you can accept that this is indeed a real problem. But even here you need to beware. Sometimes authors will self-disclose a minor problem to prove their objectivity but hide information about more serious problems.

An even more bizarre twist to the Google search. This very page appears #3 on the Google list (search terms: milk molasses enema). I'm not sure whether to be flattered or frightened. I'm already the Dear Abby of Statistics on the Internet, but I don't want to be thought of as an expert on enemas! Maybe I need to include a prominent disclaimer. (Latest update on December 15--my listing on Google has thankfully slipped to #12. Perhaps it is a function of this weblog entry slipping further to the bottom of this web page.)

Another update (December 15, 2005) I just saw an email mentioning our hospital license for UpToDate. Here's what the web site for UpToDate says:

UpToDate is specifically designed to answer the clinical questions that arise in daily practice and to do so quickly and easily so that it can be used right at the point of care. Our physician editors and authors review and update our content on a continuous basis and a new, peer-reviewed version is issued every four months. The published evidence is summarized and specific recommendations made for patient care. Independent research shows that UpToDate outperforms other resources and has a significant impact on patient care.

I thought I'd try it out on a search for "enema." There were a lot of responses that were not relevant, but an article

provided some useful guidance. This author recommended a phosphate sodium enema, or a mineral oil enema followed by a phosphate enema. He recommended against several other types of enemas, though

The use of soapsuds [34,35], tap water [36,37], magnesium (ie, epsom salt) [38-40], and herbal [3] enemas is not recommended because of potential complications (colitis, water intoxication, bowel perforation, bowel necrosis) [3,5].

He does not mention the milk with molasses enema. Here are the references that he cites:

3. Constipation and encopresis in childhood. Abi-Hanna A; Lake AM. Pediatr Rev 1998 Jan;19(1):23-30; quiz 31.

5. Constipation in infants and children: evaluation and treatment. A medical position statement of the North American Society for Pediatric Gastroenterology and Nutrition. Baker SS; Liptak GS; Colletti RB; Croffie JM; Di Lorenzo C; Ector W; Nurko S. J Pediatr Gastroenterol Nutr 1999 Nov;29(5):612-26.

34. Severe colitis induced by soap enemas. Orchard JL; Lawson R SO. South Med J 1986 Nov;79(11):1459-60.

35. Soap colitis. Pike, BF, Phillippi, PJ, Lawson, EH Jr. N Engl J Med 1971; 285:217.

36. Water intoxication following tap-water enemas. Ziskind, A, Gellis, SS. AMA J Dis Child 1958; 96:699.

37. Hyponatraemia from tap-water enema. Chertow, GM, Brady, HR. Lancet 1994; 344:748.

38. Severe magnesium toxicity after magnesium sulphate enema in a chronically constipated child. Ashton MR; Sutton D; Nielsen M. BMJ 1990 Feb 24;300(6723):541.

39. Hazards of hypertonic magnesium enema therapy. Brown, AT, Campbell, WA. Arch Dis Child 1978; 53:920.

40. Fatal hypermagnesemia caused by an Epsom salt enema: a case illustration. Tofil NM; Benner KW; Winkler MK. South Med J 2005 Feb;98(2):253-6.

Some of these references appear to be case series, based on their titles or their abstracts (which I have not included here), but I have not had a chance to check out the others.

This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Information searching.