Article Text

Download PDFPDF

Acute pulmonary oedema complicating polyethylene glycol intestinal lavage
  1. A Wong,
  2. G L Briars
  1. Department of Paediatrics, West Suffolk Hospital, Bury St Edmunds, UK
  1. Correspondence:
    Dr Briars, Department of Paediatrics, West Suffolk Hospital, Hardwick Lane, Bury St Edmunds IP33 2QZ, UK;

Statistics from

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.

Polyethylene glycol (PEG) has gained a reputation as safe and well tolerated1 method of bowel preparation for colonoscopy, colonic surgery, radiological examination, and faecal dysimpaction in intractable constipation. PEG 3350 is a chemically inert polymer of formula H(OCH2CH2)n where n = 68–84 with negligible (<0.06%) gastrointestinal tract absorption in normal subjects. This limits the potential for systemic toxicity.2 PEG causes a substantial flow of faecal water by osmosis, and because volumes required for bowel cleansing are larger than many patients can comfortably drink, it is frequently given through a nasogastric tube.

Reported side effects include transient nausea and vomiting, abdominal fullness, cramps, and anal irritation. Allergic reactions have been reported rarely. Extra care is advised in patients with impaired gag reflex, reflux oesophagitis, and those with diminished levels of consciousness.3 Pulmonary oedema in response to aspiration of hypertonic contrast material has been reported before,4 but not with intestinal lavage solutions. We report a case of acute pulmonary oedema secondary to aspiration of PEG.


A 7 year old boy with intractable constipation was admitted to hospital for nasogastric PEG (Klean-Prep, Norgine) faecal dysimpaction. Correct tube placement was confirmed by auscultation and litmus testing. PEG infusion was given by gravity feed at a rate of 15 ml/kg/h. In the course of the infusion he intermittently dozed. After 3.5 hours, he coughed, retched, and vomited approximately 150 ml of frothy fluid. Within half an hour, he became pale and unwell with intercostal recession, crackles in both his lung fields, and an oxygen saturation of 88% in air. PEG administration was discontinued and the nasogastric tube removed. Blood pressure was 110/75 mm Hg, and capillary refill time three seconds. He improved with oxygen and frusemide therapy with arterial pH 7.367, pco2 5.83 kPa, and po2 11.25 kPa. Chest radiograph (fig 1) showed extensive alveolar shadowing throughout both lung fields consistent with pulmonary oedema.

Figure 1

Chest radiograph showing extensive alveolar shadowing throughout both lung fields consistent with pulmonary oedema

Despite hydrocortisone and antibiotic therapy his condition steadily deteriorated. At 45 minutes, he required continuous positive airway pressure via facial mask and then mechanical ventilation. His chest radiograph showed notable improvement after eight hours, and he was successfully extubated after 36 hours. He made a full recovery.

Subsequent investigations have shown a normal upper endoscopy, echocardiogram, and oesophageal pH probe study with a reflux index of 4.9%, and a ganglionated rectal biopsy.


We speculate that in our case, vomiting resulted in migration of the nasogastric tube into the oesophagus, where PEG administration overwhelmed normal oesophageal peristalsis, causing aspiration. We had no clinical indication of intestinal dysmotility and the assumption that his motility was normal is supported by our subsequent investigation. This leads us to believe that the risk of this complication in our patient is no greater than that in any other patient undergoing intestinal lavage. The intense osmotic action of the PEG allowed a small aspiration to facilitate pulmonary oedema, and its molecular size prevented pulmonary absorption with a continuing osmotic action and the ongoing requirement for ventilation.

We have become aware of anecdotal reports of four previous cases in two children treated for constipation and two adults receiving PEG lavage for lithium ingestion. While information is limited, recovery was reported in all four, with 48 hours of ventilation in the paediatric cases.

Pulmonary aspiration and pulmonary oedema is an important complication of this commonly used therapy. We now advise active pumping of PEG starting at 5 ml/kg/h, and increasing by 5 ml/kg/h increments as tolerated, to a maximum of 20 ml/kg/h, to avoid vomiting. We advise rechecking the nasogastric tube position after any vomit and discontinuation of the infusion when the patient is not fully awake.