Q I‘ve been told that I need a uterine biopsy and wondered if you had any information on this procedure that may help me weigh up the pros and cons. I would particularly like to know if there is any evidence to show that biopsy is helpful and what the risks of the procedure are.- BV, Enfield

A Biopsy is an increasingly common way of obtaining tissue samples for laboratory examination. While doctors refer to it as a ‘procedure’, it’s important to remember that it is, in fact, a form of surgery and, like any surgical procedure, it carries risks (J Bone Joint Surg Am, 1996; 78: 656-63). What these are depend, among other things, on the type of biopsy performed, the reason for performing it in the first place and the skill of the surgeon.

There are several types of biopsy, ranging from those using fine needles to those requiring an open incision. While most of us associate a biopsy with the presence of a lump or tumour, this is not always the case. Any human tissue can be biopsied and this procedure is now widely used to extract all types of living tissue to examine it for all kinds of irregularities.

While biopsies can be very accurate in determining whether a mass or tissue is abnormal or cancerous, they are often performed for confused reasons.

For instance, women with an intact uterus who are on oestrogen therapy usually take an additional hormone, progestin. This has become standard practice ever since it was discovered that unopposed oestrogen causes an increased incidence of endometrial (uterine) cancer. While progestin is believed to offer protection against this possibility, it can also cause bleeding.

Ironically, vaginal bleeding is also a sign of endometrial cancer, leading gynaecologists to recommend endometrial biopsy before a woman begins oestrogen/progestin therapy to rule out the presence of an existing cancer.

This type of biopsy - where the doctor removes cells from the endometrium, or lining of the uterus - can be particularly uncomfortable, resulting in cramps and more breakthrough bleeding.

Many women might consider the transient pain caused by a biopsy to be worthwhile if the procedure itself was sound. Unfortunately, there is no scientific evidence to support this practice. In 1996, the US Preventive Services Task Force reported that endometrial biopsy has never been adequately assessed for its ability to identify endometrial cancer. This fact foreshadowed several damning studies on the ‘miracle’of biopsies.

One involved nearly 3000 biopsied women. The researchers concluded that the incidence of endometrial cancer found was too minuscule - at 0.07 per cent - to justify the routine use of this procedure (Am J Obstet Gynecol, 1997; 176: 377-80).

This conclusion has been echoed in other studies of the predictive value of biopsy, including one on breast biopsy. According to researchers, the majority of unnecessary biopsies are carried out on ‘serendipitous’ tumours - those found during magnetic resonance imaging (MRI), but which didn’t show up on mammography or during breast examination. Such tumours are unlikely to be malignant.

Indeed, MRI exams can result in high levels of false-positives (indicating cancer where there is none) because the machine is so sensitive that it may be picking up potential ‘abnormalities’ that represent nothing more than the normal terrain of the breast (J Natl Cancer Inst, 1998; 90: 1792-800).

Recently, a Canadian study found that preoperative liver biopsy is unnecessary for a diagnosis of liver cancer and that, for patients with smaller tumours, the risk of a false-positive finding is unacceptably high (Ann Surg, 2001; 234: 206-9).

Whether or not you should agree to a biopsy depends on several things. You need to be clear in your mind as to what you would do if the result reveals a malignancy. If you would not under any circumstances consider conventional treatment, then there is little point in having a biopsy.

If you and your physician are convinced that cancer is present, then you should also consider, as far as possible, what the real risks are that the biopsy itself will cause spreading (or ‘seeding’, the medical term) of the cancer.

Most doctors underestimate the risk of tumour seeding due to biopsy. Yet, there is ample evidence that this is a genuine possibility. Under normal circumstances, cancer in organs or specific tissues can, to some extent, be ‘walled off’ from the rest of the body, allowing the body to deal with it locally. However, once that natural barrier is breached by a needle or scalpel, cancer cells may leak out and spread.

It is known that biopsy of undescended testicles - assumed to increase a man’s risk of developing testicular cancer later in life - often reveals nothing of clinical use and, instead, increases the risk of cancer. In such cases, a biopsy is routinely performed at the time of orchiopexy (an operation to secure the testis within the scrotal sac), ostensibly to help the doctor judge the risk of developing cancer. But, in a study of more than 120 men with undescended testicles for up to 45 years, the researchers concluded that biopsy was 'a stronger risk factor for testicular cancer than any factor previously identified' (BMJ, 1997; 314: 1507-11).

There are case reports of liver cancer being seeded by biopsy, with an overall risk of around 0.5-2 per cent (Gut, 1999; 45: 626-7; Ital J Gastroenterol, 1995; 27: 473-8; J Hepatol, 2001; 35: 254-8). Indeed, in one study, the risk was as high as 5.1 per cent (Liver Transpl Surg, 2000; 6: 67-72). The literature also includes reports of seeding of prostatic cancer from needle biopsy (J Surg Oncol, 1987; 35: 184-91).

There is evidence that tumour seeding can occur after both conventional and fine-needle biopsies, though the risk appears to rise according to how invasive the procedure is. Thus, large-needle biopsy is considered more risky than fine-needle procedures (Acta Cytol, 1995; 39: 449-52). In one large-needle study, the overall risk of seeding was 0.012 per cent, with an average interval between the procedure and metastasis of 2.6 months (J Thorac Imag, 1998; 13: 2-6).

There are other problems to consider with biopsy. While the less invasive fine-needle biopsy may be less likely to cause the uncomfortable and risky adverse effects associated with open surgery, there are problems because of the small size of the tissue sample that is removed. Because of this, fine-needle biopsy is more subject to diagnostic errors, particularly in the grading of tumours. In addition, they may be more likely to miss the target area so that, with fine-needle biopsy, the surgeon may need several attempts to obtain a useful sample.

Indeed, a team of US researchers has concluded that fine-needle aspiration biopsy of the breast frequently does not collect enough tissue to make a useful diagnosis. An analysis of 377 women revealed that, overall, more than a third of the biopsies were inadequate (Cancer, 1998; 82: 679-88). When the same team updated their data last year, they found that time had not improved the detection rates (Radiology, 2001; 219: 785-92).

Even more revealing was a 1999 study from Johns Hopkins involving more than 6000 patients. In this study, the researchers found that 1-2 per cent of those referred to larger medical centres for treatment following a biopsy arrived with a totally wrong diagnosis. Often, the diagnosis was that cancer was present when it was not, leading to inappropriate treatments such as surgery and chemotherapy (Cancer, 1999; 86: 2426-35). The conclusion of the researchers? If your biopsy shows cancer, get a second opinion on the results.

Some biopsies are downright dangerous. A review of the literature on biopsy complications in both the US and Europe suggests that abdominal procedures carry the risk of death (commonly due to liver haemorrhage and inflammation of the pancreas) in 0.031 per cent of cases (Radiology, 1991, 178: 253-8). Liver biopsies that inadvertently puncture the gallbladder can kill about 0.1 per cent of patients.

Finally, while many physicians believe that ultrasound or some other imaging technique to guide the needle makes the procedure safer and less likely to cause adverse effects, other evidence suggests the contrary.

According to data from Lincoln County Hospital, serious bleeding occurs in 2.5 per cent of guided biopsies compared with 1.6 per cent of those where the physician ‘felt’ the needle in. Pain is experienced by 25 per cent of patients who had non-guided biopsies compared with 22 per cent when the needle was guided by ultrasound. These results suggest that information from a national reporting scheme is needed to assess the two techniques (BMJ, 1994; 309: 1455-6).