JC: Do all VTE patients need a CT for cancer?

Virchester has a long history of research in the ambulatory management of venous thromboembolism. Several MDs and PhDs have been completed here looking at both therapeutic and diagnostic strategies in this rather tricky group of patients. In summary, we have got a lot better at risk stratifying, investigating and treating patients with potential VTE disease, but there are still many questions remaining. One of these is how hard should we look to determine the cause of the clot. For some patients, this is easy: the patient with recent immobility, pregnancy, trauma etc. has a provoked event and it’s pretty obvious why they have a VTE, but for many it is unclear. These unprovoked clots are a worry as many studies have shown that a significant proportion (5-10%) will have a cancer diagnosis within a year. That’s a significant proportion and clearly it’s not something we can ignore so what’s the best approach?

Clearly all these patients need a good history and examination to look for any clues to an underlying cancer. At St.Emlyn’s, we are going to presume that you know how to take a history and how to examine a patient but would suggest that you take particular care in the abdominal examination, breast exam in women over 50 and that you should take note of obvious risk factors such as smoking in the history. Routine bloods and chest X-ray are easy, low risk investigations and we would advocate it it in all patients with an unprovoked VTE.

Is that enough?

If you don’t find anything on this initial screen the question then arises as to whether a watch and wait policy is best, or should we proceed to an abdominal + pelvic CT scan. Clearly, this latter option  has implications in terms of cost and radiation risk to patient so it’s not something that we should routinely do without good reason.

This month we have an RCT to answer this question from a Canadian multicentre study published in the New England Journal of Medicine. It’s currently open access so please follow this link and read the paper for yourself.

What kind of paper is this?

This is an open label randomised controlled trial. Open label means that no blinding took place in the study for obvious reasons. Patient with unprovoked VTE were randomised to a history/exam/CXR strategy or to a strategy that additionally included an abdominal/pelvis CT scan.

Tell me about the patients.

Patients were those with unprovoked DVT attending VTE clinics in one of nine Canadian centres. Here in Virchester, we don’t have such dedicated VTE clinics and so patients can turn up in a variety of acute settings, in ED or acute medicine review clinics, on the wards or in haematology/anticoagulation settings. To some extent, this is a local problem and as a reader you need to understand the local referral patterns to consider where this patient population sits in your health care system. At St.Emlyn’s, many patients will initially be diagnosed in the ED so this is a relevant group of patients to us.

All patients had a pretty good screening similar to that we conduct here. Routine bloods, a general exam and a chest Xray were performed for all. Sex specific investigations included breast examination, mammography, or both were performed in women older than 50 years of age, then cervical smear testing and a pelvic examination in women 18 to 70 years of age. Men over 40 got a prostate exam and a PSA.

For the control group, that was it. In the intervention group, patients had a CT. This included a virtual colonoscopy and gastroscopy with a biphasic CT of the liver.

Roughly two thirds of patients had DVT, a third had PE, with about 10% having both. The groups were well matched at baseline with recruitment taking place over 6 years.

What about outcomes?

The main aim of the study was to determine the diagnosis of cancer within the first year following diagnosis. Let’s break that down though.

1. Similar numbers of patients were referred for additional testing after limited or limited+CT  (roughly 14% in both)
2. 3.9% of patients in the trial had a cancer diagnosed in the year following diagnosis (note – less than in past studies)
3. There was no statistical difference between the investigation groups: 3.2% in the limited group vs 4.5% in the CT group (p=NS)
4. Both groups included patients who were initially screened as clear but who later had a diagnosis of cancer. However, there was no difference between the groups (4 missed in limited strategy vs. 5 in the CT group). So this tells us that CT does not guarantee a diagnosis. Both strategies miss cancers.
5. There was no difference in the time it took to find those missed cancers.
6. There was no difference in mortality between the two groups.

What does this mean?

At face value, the use of CT for screening in patients with unprovoked VTE is not justified. There are a few caveats though before you abandon this as an idea for all patients.

1. These patients were in a specialised VTE clinic presumably with physicians who are highly capable and experienced in looking for occult cancers. Patients in the non-CT group had a structured assessment. Ask yourself whether your local service delivers a comparable ‘routine’ service. For example, many emergency physicians may question their ability to perform screening examinations of the breast or pelvis.
2. The study was powered for a higher VTE incidence and so may lack power to detect small differences in detection rates.
3. The overall incidence of cancer for those patients with a negative screen was similar to that in the general population. This reassures me that both approaches were effective.

Final thoughts.

In summary, this is a good trial in a group of patients who are similar to those I see in the ED. The results are good enough to base practice on, so long as your ‘routine’ and ‘limited’ strategy is as good as it was in this trial. Perhaps the main outcome for this trial is that although the diagnosis of VTE is well within the capabilities of the emergency physician, the subsequent screening and follow up may not be.

vb

S

Selected references from the real Virchester.

The Anticoagulation of Calf Thrombosis (ACT) Project: Results From the Randomized Controlled External Pilot Trial. Daniel Horner, MD; Kerstin Hogg, PhD; Richard Body, PhD; Michael J. Nash, MD; Trevor Baglin, PhD; Kevin Mackway-Jones, MA

Single whole‐leg compression ultrasound for exclusion of deep vein thrombosis in symptomatic ambulatory patients: a prospective observational cohort study

Diagnosis of pulmonary embolism with CT pulmonary angiography: a systematic review.  K Hogg, S Carley, B Foex, K Mackway–Jones – Emergency medicine …, 2006 – emj.bmj.com