Non-occlusive mesenteric ischaemia associated with anorexia nervosa

Abstract

Non-occlusive mesenteric ischaemia (NOMI) is a life-threatening condition that requires emergent intervention and anorexia nervosa is a chronic eating disorder that requires careful medical and nutritional management. A 54-year-old woman with a history of anorexia nervosa and undergoing chronic haemodialysis developed abdominal pain and called an ambulance. On arrival, she was in shock and abdominal examination was consistent with diffuse peritonitis. Computed tomography scan suggested ischaemia from the distal ileum to the ascending colon. Emergency laparotomy revealed NOMI from the distal ileum to the transverse colon. The treatment strategy included staged operations and careful medical management to optimise nutritional support and electrolyte management with survival of the patient. NOMI and anorexia nervosa are both difficult to manage. Meticulous interdisciplinary management can result in a good outcome.

Background

Non-occlusive mesenteric ischaemia (NOMI) is a catastrophic abdominal emergency. Acute mesenteric ischaemia is generally classified into acute mesenteric arterial thrombosis, acute mesenteric arterial embolus, acute mesenteric venous thrombosis and NOMI.¹ NOMI was first described as a complication of cardiac surgery in 1958.² NOMI is defined as acute mesenteric ischaemia without vascular occlusion. The aetiology of NOMI remains unclear. One hypothesis suggests that NOMI results from vasospasm to maintain circulation to vital organs.

Anorexia nervosa is an eating disorder which has a high mortality rate.³ Based on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5),⁴ the diagnosis of anorexia nervosa is defined as symptoms consistent with restriction of energy intake accompanied by significant weight loss, fear of gaining weight or of becoming fat, and disturbance of recognition or evaluation of a patient’s own body weight or shape. Anorexia nervosa is associated with disorders of the musculoskeletal system, the reproductive system, the central nervous system and growth. There are few effective treatment strategies for patients with anorexia nervosa.⁵

Anorexia nervosa can lead to hypovolemia, which may cause inadequate systemic perfusion. We present a patient with NOMI associated with anorexia nervosa.

Case presentation

A 54-year-old woman with anorexia nervosa and undergoing chronic haemodialysis developed abdominal pain and vomiting. One day before admission, she had a usual haemodialysis session that was terminated due to hypotension and insufficient dialysis. She then developed nausea and bloating, which continued through the night. On the day of admission, her family noted that she was having severe abdominal pain and called an ambulance. The patient had an 18-year history of anorexia nervosa, hypothyroidism, chronic hepatitis of unknown aetiology and had been undergoing haemodialysis for 1 year. She usually drank only a nutrient drink as energy and her body mass index (BMI) was 14.4 kg/m². She was classified as a restricting and extreme (BMI <15 kg/m²) type of anorexia nervosa based on DSM-5.⁴ She started haemodialysis for renal insufficiency caused by abuse of diuretics.

On arrival, the patient was in acute distress with severe abdominal pain. She was conscious, her blood pressure could not be measured, the pulse was 105/min and she was tachypneic. Physical examination was consistent with diffuse peritonitis.

Investigations

Laboratory studies showed lactic acidosis (lactate 14 mmol/L) and elevated hepatic enzymes (Aspartate transaminase 318 U/L, Alanine transaminase 232 U/L). After fluid resuscitation, a contrast-enhanced CT scan was obtained and showed intramural gas in the ileum and ascending colon without occlusion of the celiac or superior mesenteric arteries (figure 1).

Figure 1.

Contrast-enhanced CT scan of the abdomen. Red arrowheads show intramural gas in the ascending colon.

Differential diagnosis

Based on the history, physical examination and CT scan findings, the patient was diagnosed with NOMI requiring emergent laparotomy.

Treatment

At operation, the intestine from the distal ileum to the transverse colon appeared grossly necrotic, although there was no obvious perforation. The necrotic bowel including 40 cm length of the distal ileum and the right colon were resected (figure 2). The patient was haemodynamically unstable and no anastomosis was performed. The stapled ends of the bowel remained and we performed a temporary abdominal closure. Postoperatively, it was still difficult to maintain her blood pressure without catecholamine support. Broad-spectrum antibiotics were administered. A second-look operation was performed on hospital day two to evaluate the remaining bowel. There was 20 cm more necrotic bowel distally, which was resected. On the fourth hospital day, a third-look operation revealed no further necrotic bowel and an end ileostomy was made. There was approximately 150 cm of small intestine remaining.

Figure 2.

Surgical specimen with necrotic colon and ileum from the first and second operations. (A) The distal ileum has several ischaemic skip lesions and the colon from caecum to transverse has ischaemic mucosa, resected at the first operation. (B) Approximately 20 cm of necrotic distal ileum was resected at the second-look operation. White arrowheads identify the distal end of the specimen.

Outcome and follow-up

After the third procedure, enteral nutrition was started and changed to oral intake after 2 weeks of enteral nutrition. The patient had difficulty eating solid food because she had not eaten solid food for a long time. She succeeded to take in 1000 kcal/day orally with a low residue diet without parenteral nutrition. Her metabolic condition was monitored by measuring serum electrolytes including phosphate and magnesium, liver function tests and cardiac function to monitor refeeding syndrome. After an inpatient stay of 3 months, she was discharged to a nursing home.

Discussion

We present here a patient with NOMI associated with anorexia nervosa. NOMI is reported to have 60%–100% mortality.¹ We managed the patient successfully with a bowel resection followed by two repeat laparotomies and careful nutrient support. There is a report of a patient with NOMI and anorexia nervosa. In this paper, the authors performed a single laparotomy several days after the onset of abdominal symptoms and the patient did not survive.⁶ To our knowledge, ours is the first report of survival in a patient with NOMI associated with anorexia nervosa.

Haemodialysis was reported as a risk factor for the development of NOMI and most of the patients diagnosed with NOMI had abdominal pain more than 1 day before admission.⁷ The present patient had abdominal pain which started the day before admission. Judging from the fact that the abdominal pain started a few hours after haemodialysis, it may have been related to hypovolemia and mesenteric hypoperfusion. It is difficult to definitively attribute the development of NOMI only to haemodialysis, but it must be considered.

The management of patients with NOMI includes optimisation of volume status and assessment of intestinal viability. The management of the intestine depends on the presence of necrosis. In patients without necrotic intestine, placement of an arterial infusion catheter by interventional radiology and infusion with a vasodilator may be sufficient.⁸ Exploratory laparotomy and bowel resection are required for patients with necrotic bowel.⁹ Repeat laparotomy (second look, third look and so on) and temporary abdominal closure are important for the management of many patients especially those who are physiologically unstable, as was done in the present patient.⁹ Early detection and intervention are important in the management of patients with NOMI.⁸ The present patient had signs of diffuse peritonitis on physical examination at the time of presentation and needed emergent exploratory laparotomy. The bowel was left with the ends stapled and a temporary abdominal closure was performed because the patient was haemodynamically unstable during the operation. Ileostomy was not performed until the third operation, by which time she had become haemodynamically stable.

Regarding anorexia nervosa and hypothyroidism, the patient was on medications including benzodiazepine, levothyroxine and multivitamin. There is no direct evidence implicating medications for anorexia nervosa as an aetiology of NOMI.¹⁰ A patient with anorexia nervosa sometimes needs urgent hospitalisation for nutritional or cardiovascular problems based on his/her clinical condition, cardiovascular status or electrolyte disturbance.⁵ Refeeding syndrome is a critical complication of anorexia nervosa. Careful physical examination and determination of serum levels of phosphate, magnesium, potassium and glucose are important.¹⁰ We began repletion of electrolytes just after the first operation. In the early stage of the hospitalisation, the patient experienced several episodes of hypoglycaemia without clinical symptoms. It is unclear whether asymptomatic hypoglycaemia needs treatment. It is suggested that undernutrition affects glucose metabolism and resting energy expenditure to limit the loss of body mass.¹¹

It has been suggested that the length of small intestine remaining after resection should be greater than 180 cm to absorb enough nutrients for long-term survival.¹² However, the present patient may have had a very low resting energy expenditure because of anorexia nervosa. In addition, intestinal adaptation is expected. We did not use indirect calorimetry; however, judging from her stable serum glucose level and stable body weight during the hospital stay, the patient’s metabolic needs can be considered maintained while she needed continuous intervention for anorexia nervosa. Long-term follow-up is essential.

Learning points.

• Non-occlusive mesenteric ischaemia (NOMI) is difficult to manage and may require multiple operations and anorexia nervosa requires interdisciplinary intervention.

• It is plausible that the development of NOMI in this patient is related to the history of anorexia nervosa, and we need to consider mesenteric ischaemic in the differential diagnosis of a patient with anorexia nervosa and new onset abdominal pain.

• When a patient has multiple conditions especially a catastrophic illness such as NOMI, careful and intensive multidisciplinary management is essential.