Arab Board Surgery Part 1 study materials
Arab Board Surgery Part 1 study resources

Milky Drain, Drowsy Patient: Refeeding After Oesophagectomy

Arab Board Surgery Part 1 · Put your knowledge to work in context.

All case studies

Milky Drain, Drowsy Patient: Refeeding After Oesophagectomy

A 58-year-old man is on postoperative day 6 after an Ivor Lewis oesophagectomy for adenocarcinoma. A persistent low-grade fever developed around 24 hours after enteral feeding was started, and the chest drain has produced 1.4 L/day of milky fluid since then. His only other fluid is 1 L of 5% dextrose daily. He is drowsy but rousable. Weight is 68 kg (78 kg preoperatively). Sodium 131 mmol/L, potassium 2.8 mmol/L, phosphate 0.38 mmol/L, magnesium 0.48 mmol/L, corrected calcium 1.90 mmol/L, glucose 4.1 mmol/L, albumin 22 g/L. The abdomen is soft, he is passing flatus, and there is no surgical emphysema or anastomotic leak on contrast study.

1. Explain what is happening metabolically and identify the most time-critical abnormality. Outline your correction plan for the first 24 hours.

Read the worked answer

There are two parallel processes.

1. Chyle leak. Milky drain fluid appearing after enteral feeding, at high volume, is thoracic duct injury. The duct runs between the aorta and azygos vein in the right chest and is at risk during oesophageal mobilisation. High-volume losses produce hypovolaemia and loss of protein, lymphocytes and electrolytes - hence the albumin of 22 and the hyponatraemia.

2. Refeeding syndrome. After six days of essentially no nutrition (1 L of 5% dextrose provides about 170 kcal), the body has depleted intracellular phosphate, potassium and magnesium stores. Introducing feed triggers an insulin surge that drives these electrolytes intracellularly. The phosphate of 0.38 mmol/L is the most time-critical value - severe hypophosphataemia causes respiratory muscle weakness, cardiac failure and rhabdomyolysis, and it can kill within 24 hours. The drowsiness should be assumed to be metabolic until proven otherwise.

First 24 hours:

  • Thiamine and multivitamins before any carbohydrate to prevent Wernicke's encephalopathy or precipitation of beriberi.
  • Replete electrolytes intravenously: potassium and magnesium replacement and, critically, intravenous phosphate. Recheck every 6–12 hours because intracellular shifts continue as feeding escalates.
  • Restrict and rationalise fluid. Stop the 1 L of 5% dextrose - it is worsening the hyponatraemia without providing meaningful nutrition.
  • Start nutrition at 10–20 kcal/kg/day, i.e. around 700–1000 kcal, and increase slowly over 4–7 days while monitoring phosphate; do not chase full calculated requirements in the first 48 hours.

2. How would you feed this patient, and how would you manage the chyle leak? Discuss the anatomy and the thresholds for reoperation.

Read the worked answer

Route. The gut works and he is passing flatus, so enteral access is preferable in principle - but high-volume chyle leak changes this. Options:

  • Enteral feeding with a fat-modified formula (medium-chain triglycerides, which are absorbed directly into the portal system and bypass the lacteals), plus a low-fat diet. This often reduces flow within days.
  • Nil by mouth with total parenteral nutrition if drain output remains above roughly 1 L/day, or if electrolytes and albumin cannot be maintained enterally. TPN is also the cleaner choice while he is in negative fluid balance and needs careful refeeding.

Managing the leak conservatively: keep the drain on low-pressure suction and record volumes; keep him nil by mouth if output is high; replace losses with crystalloid; monitor lymphocytes and albumin; and give octreotide only as an adjunct, recognising the evidence is weak. There is no role for prophylactic antibiotics in an uncontaminated chyle leak.

Thresholds for reoperation:

  • Output persistently above 1–1.5 L/day for 5–7 days despite dietary measures;
  • Progressive nutritional and electrolyte deterioration despite maximal support;
  • A loculated collection, empyema or sepsis not controllable by drainage;
  • Any suspicion of conduit ischaemia or anastomotic failure.

The operation is thoracic duct ligation, best approached on the right side through the chest, or by mass ligation of the tissue between the aorta, azygos vein and oesophagus. Before ligation, cream given down the nasogastric tube can help identify the leak.

3. On day 8 the drain output falls but he becomes tachycardic with a new left-sided pleural collection. How do you prioritise, and what does this teach you about interpreting a falling drain volume?

Read the worked answer

A falling drain volume is not evidence of resolution. It may mean the leak has sealed, but it may equally mean the drain has blocked, migrated or become loculated, with chyle now accumulating in the chest and causing a pleural collection with systemic inflammation. The new tachycardia and left-sided collection make the second explanation at least as likely as the first.

Priorities:

  1. Assess the patient, not the number. Oxygen saturations, respiratory rate, work of breathing, lactate and white cell count. A blocked drain in the presence of an active leak is a surgical problem, not a reason to relax.
  2. Image. CT of the chest with contrast to define the collection, its size and whether it is drainable, and to look for mediastinal contamination.
  3. Drain and send. If there is a significant collection, insert an image-guided drain and send fluid for triglycerides, microbiology and cell count. A triglyceride-rich fluid confirms chyle; a low pH or positive Gram stain suggests empyema requiring more aggressive intervention.
  4. Support and reassess. Continue nutritional support, correct electrolytes, and re-evaluate for operative thoracic duct ligation if drainage persists or sepsis develops.

The broader lesson is that drain output must always be interpreted alongside the patient's physiology and imaging; a fall in output with a deterioration in the patient is a red flag, not reassurance.