“Could I ask you for some advice about a patient who came in last night please?” asked the very polite haematology junior doctor.
“Go one then” muttered the less polite Microbiologist wondering when he was going to finish looking at the hundreds of emails he had accrued whilst on holiday last week.
“We have a sixty year old lady who came in with febrile neutropaenia last night following her chemotherapy two weeks ago for leukaemia. There was no obvious cause of the fever so she was started on Piptazobactam but she has worsened overnight. She has now started to complain of abdominal pain and we want to make sure we are covering gastroenteritis.”
The Microbiologist pricked up his ears, not only was this a nice potted summary of the problem but the abdominal pain was ringing alarm bells.
“Where’s the pain?” he asked followed rapidly by “how long has she been neutropaenic?”
“Erm…” stalled the junior doctor “worse on the right side and she has been neutropaenic for about two weeks.”
“Okay, this might be typhlitis, this needs immediate action as it’s actually an emergency and after telling your own Consultant here’s what I want you to do….”
Crikey thought the junior doctor, I only wanted to know what antibiotic to give for gastroenteritis…
Typhlitis, or to give it it’s other name of neutropaenic enterocolitis, is severe life-threatening infection of the caecum in a patient whose immune system isn’t working properly. The term typhlitis is derived from the Greek word “typhlon” meaning caecum, and has nothing at all to do with Teflon and non-stick surfaces on frying pans!
Typhlitis is more common in neutropaenic patients with an underlying haematological malignancy rather than patients who are neutropaenic with other forms of cancer. This might be because the white blood cells in haematology patients are not functioning very well because of the cancer, even before the patient becomes neutropaenic. We often use the descriptive term “functionally neutropaenic” to describe what is going on as although the patient might have neutrophils, they don’t work properly, so for all intents and purposes the patient might as well be neutropaenic.
In typhlitis, bacteria and fungi penetrate into the wall of the caecum causing tissue destruction and necrosis. Infection can then spread to the adjoining ileum or ascending colon. The reason why it starts in the caecum is thought to be because the caecum is at a “watershed” in terms of blood supply (an area of the body that is supplied by separate arteries which neither supply very well); the caecum is at the end point of the blood vessels so has a poor blood supply compared to the rest of the bowel and is prone to ischaemia.
What bacteria and fungi cause typhlitis?
There is no single cause of typhlitis. Typhlitis is a type of synergistic infection where multiple different bowel bacteria cause damage, a bit like synergistic gangrene in necrotising fasciitis. It is always a mixture of bacteria which cause the infection; Gram-negative bacilli, Gram-positive cocci, anaerobes, facultative anaerobes and Candida spp.
How does typhlitis present?
The most common presentation of typhlitis is a patient with a neutrophil count < 0.5 x 109 per ml AND abdominal pain, usually in the right lower quadrant. It is essential that the differential diagnosis of such a patient includes typhlitis so that it isn’t missed. It most commonly occurs two weeks or more after chemotherapy when the neutropaenia is most pronounced.
Other symptoms include:
- Abdominal distension and tenderness
- Watery or bloody diarrhoea
Typhlitis is usually confirmed using a CT scan of the abdomen that shows bowel wall thickening due to oedema, mesenteric stranding (white streaks on radiology in the mesenteric part of the bowel) due to inflammation, dilatation of the bowel and pneumatosis coli (gas in the bowel wall produced by the bacteria themselves).
How is typhlitis treated?
The treatment of typhlitis includes a combination of supportive care, antibiotics and surgical assessment.
Supportive care of the patient should include keeping them “nil by mouth” with a nasogastric tube in place to rest the bowel. IV fluids are needed as well as nutritional support if the condition doesn’t settle within a few days (and it often doesn’t!). Consideration should be given to trying to help the neutrophil count to recover using drugs such as Granulocyte Colony Stimulating Factor (GCSF), but this is up to the Consultant Haematologist looking after the patient.
Antibiotics for typhlitis have to be very broad spectrum, covering all of the common bowel bacteria and fungi, as well as cidal, that is they can kill the bacteria and fungi without the patient having a working immune system.
My personal preference is for IV Meropenem PLUS IV Metronidazole PLUS IV Caspofungin. My reasons for this are:
1. Meropenem – about as broad-spectrum as it gets with excellent cover of the common bowel flora
2. Metronidazole – this is one of the few occasions I double the anaerobic cover by using Metronidazole with Meropenem as the anaerobes are a major part of the problem in typhlitis and I want to be sure they are killed (Dead, Dead, Dead!!!)
3. Caspofungin - a broad spectrum antifungal with excellent cidal activity against Candida spp., it is also active against other fungi such as Aspergillus spp. which can just occasionally be part of the problem.
There is no specific duration of treatment for typhlitis; it depends on how long it takes for the neutrophil count to recover and the patient to improve. I’m also not in a hurry to convert to oral antibiotics in these patients as the bowel is often not working well even when the symptoms settle and I am worried the patient won’t absorb the oral doses. I tend to continue IV treatment for 5 days after the typhlitis has settled and then stop.
Although an urgent surgical review is necessary, ideally surgery should NOT be performed on patients with typhlitis as they often do not heal well afterwards because of the underlying blood problems. However, if the patient perforates their bowel or continues to deteriorate despite maximal medical therapy then colectomy may be needed. It is important that the Surgeon removes the entire affected bowel, and this can be difficult as submucosal necrosis can be present without obvious superficial signs. The Surgeon should error on the side of more aggressive surgery as patients tend to die if the entire necrotic bowel isn’t removed.
The mortality of typhlitis has been described as being about 50% but with aggressive supportive care and antibiotics the mortality is probably a lot lower. From personal experience most patients survive as long as the diagnosis is made quickly and good clinical care implemented.
So the Haematologists made the patient nil by mouth and started supportive care. IV Meropenem PLUS Metronidazole PLUS Caspofungin were started quickly and the urgent CT scan confirmed the diagnosis of typhlitis. The surgeons came and saw the patient and were reassured that surgery wasn’t indicated at this time.
The patient slowly recovered over the next week and eventually was able to go home. Plans were made to keep a close eye on the patient during the next cycle of chemotherapy, as having had typhlitis once they were at risk of further episodes.
Oh and while we’re at it most gastroenteritis is self-limiting, even in neutropaenic patients… so the answer to the junior doctors original question about what antibiotic to give for gastroenteritis is “none”.