Changing paradigms in the investigation of an elevated PSA level


Correspondence: Roger Kirby, The Prostate Centre, London W1G 8GT, UK.



3 Tesla




acute urinary retention

Until recently, a raised PSA level has usually prompted referral of the patient for a TRUS-guided biopsy of the prostate, performed under local anaesthetic, with antibiotic cover to exclude a diagnosis of prostate cancer. Several recent developments, however, have called this practice into question. Firstly, data from the European Randomised Study of Prostate Cancer Screening suggest that, although a 23% reduction in prostate cancer mortality can be achieved by PSA screening, this is at the cost of the ‘over-diagnosis’ of clinically insignificant cancers by ∼50% [1]. Secondly, there are concerns about the rising incidence of serious infections, usually Escherichia coli septicaemia, that can occur within a few days of TRUS biopsy in spite of appropriate antibiotic prophylaxis [2]. Thirdly, the development of multiparametric MRI of the prostate, especially with larger strength 3-Tesla (3T) magnets [3].

An alternative approach, and one that we have recently adopted, is to respond to the raised PSA level, or one that is rising progressively over time, by requesting a 3T MRI scan with diffusion-weighted sequences and gadolinium enhancement (Fig. 1) for post-PSA test and pre-biopsy risk stratification. There is now increasing evidence that this technology can identify clinically significant prostate cancers, not only in the peripheral zone of the prostate but also in the anterior gland [4]. Significantly, it is the latter group of tumours that are liable to be missed by TRUS-guided biopsies as a result of the direction of travel of the automated biopsy needle through the transrectal probe. As a consequence, these cancers tend to be diagnosed later on repeat biopsy when the PSA level continues to rise in spite of the negative initial TRUS-guided biopsy.

Figure 1.

Photomontage of typical series of multiparametric 3T MRI with gadolinium enhancement images of the prostate gland. A is a T2-weighted image showing a diffuse area of low signal in the left apex (arrow). On B, an apparent diffusion coefficient map, this area is seen to be of restricted diffusion. C is a diffusion-weighted image, and confirms that this area is of high signal indicating a higher cellular density, as may be seen with prostate cancer. Finally, this area shows avid early phase vascualirty (D) in keeping with neovascularised tissues. Such concordant changes as seen in all sequences of a multiparametric series, are virtually diagnostic of prostate cancer, and more often seen with high grade or clinically significant cancers. A single directed biopsy would be sufficient for diagnosis, and taken with the rest of the MRI findings, also enough for treatment planning.

A high-quality 3T MRI scan, reported on by a radiologist who is experienced in evaluating these images, can therefore not only identify an ‘area of interest’ in the peripheral zone, which can be sampled by the TRUS-guided route, but also recognise the less common anterior and central zone cancers, which are better approached via the transperineal (TP) route. Prostate biopsy would become a more intelligently directed procedure, especially with the use of MRI-ultrasonography fusion of images [5]. TP biopsy is already increasingly used by clinicians, mainly because of the almost negligible post-biopsy infection risk, and because some patients prefer general to local anaesthesia while the biopsies are being taken. TP prostate biopsy also enables a larger number of biopsies to be taken (some advocate that as many as 70 biopsies should be taken to accomplish so-called ‘prostate mapping’). The chief disadvantage of TP biopsy to the patient, however, is a significant risk of post-biopsy acute urinary retention (AUR). This is probably a direct result of the track of the needles through the perineal body, which perturbs the external urinary sphincter, and the more extensive sampling of the central zone of the prostate, which is closer in proximity to the urethra. As a consequence it is usually wise to leave a catheter in situ for a period after TP biopsy, and to warn the patient about the risk of AUR.

Prostate biopsy, accomplished by whichever route, frequently results in haematuria and haemospermia, which may be both alarming and persistent.

Patients therefore need to be warned about these sequelae in advance. They should also receive written information about the risk of post-biopsy sepsis after TRUS biopsy and how to respond to such a scenario, since delay in treatment of E. coli septicaemia can have grave consequences. The development of rigors and pyrexia, often accompanied by flu-like symptoms and sometimes confusion, requires urgent treatment with i.v. antibiotics. Unfortunately, recently, there has been a rise in the incidence of E. coli species resistant to both gentamicin and ciprofloxacin, the very two antibiotics most commonly deployed as prophylactics before prostate biopsy. There is therefore a strong case to be made for using amikacin as an alternative prophylactic and meropenem, rather than gentamicin, as first-line treatment of established post-biopsy septicaemia.

Concerns will undoubtedly be raised, especially in these recessionary times, about the increased costs associated with using multiparametric MRI (ideally 3T) as the first-line investigation of an elevated PSA level. Its value has yet to be proven in large scale, controlled studies. In particular its true negative rate for clinically significant cancers when used outside the specialist centres, needs to be established, and the results of the PROMIS trial ( are keenly awaited. Other radiological screening investigations such as mammography and CT colonography, have shown the importance of defined technical and reporting standards. These issues are being addressed [4] but there is the separate issue of the logistics and costs of this strategy. Quantification of the cost to the UK NHS of such a move is difficult since there are no published figures available for the number of prostate biopsies performed annually in the UK; however, it can be estimated: if there are in the order of 40 000 new cases of prostate cancer diagnosed in the UK each year, and assuming that around 1 in 3 TRUS-guided biopsies are positive for cancer, there must be >100 000 prostate biopsies being performed to establish those diagnoses. The provision of >100 000 prostate MRI scans would be a substantial burden to the NHS economy. In addition, more widespread use of 3T MRI in this situation would require not only the acquisition of a great deal of expensive technology but also the training of large numbers of radiologists in the interpretation of the images, as even modern studies continue to report a striking interobserver variability [6].

The potential benefits, however, of adopting this new paradigm are as follows:

  1. More accurate pre-biopsy localisation of suspicious areas in the prostate, allowing more accurate sampling and informing the decision about whether a TP or TRUS-guided approach is most suitable.
  2. Avoidance of unnecessary biopsies in those men in whom only benign features are present on MRI.
  3. A less invasive means of following up those men whose ‘low risk’ disease is managed by active surveillance, rather than initial surgery or radiotherapy.
  4. Improved preoperative staging of those cancers deemed suitable for surgery in terms of the presence or absence of capsular invasion and penetration.
  5. A reduction in the numbers of ‘incidental’, low-risk prostate cancers diagnosed which are unlikely to progress within the individual's natural lifespan.

It remains to be seen how quickly this new, and not inexpensive, MRI technology will be introduced internationally, and to what extent urologists change their practice in response to these latest developments.

Conflict of Interest

None declared.