Detection of bladder dysplasia using 5-aminolaevulinic acid-induced porphyrin fluorescence


Dr M. Colombel, Service d’Urologie et Chirurgie de la Transplantation, Hôpital Edouard Herriot, 69437 Lyon, Cedex 03, France.



To report results of a clinical investigation on the detection of bladder dysplasia and in situ carcinoma by using fluorescence induced by 5-aminolaevulinic acid (ALA).


The study included 50 patients with a primary bladder lesion, who had a bladder instillation of 50 mL of 3% ALA solution ≥ 1 h before transurethral resection of the tumour. Random biopsies were taken using white-light cystoscopy, then using blue light to induce fluorescence; positive zones were noted and biopsied. The primary lesion was then resected. The frequency of dysplasia detected by ALA-induced fluorescence was evaluated, as was the risk of recurrence with a follow-up of ≥ 2 years.


In all patients the tumours were positive; in 21 fluorescence distant from the tumour was detected. The pathological report of the biopsies showed 11 cases of dysplasia, six of carcinoma in situ and four of inflammatory lesions. In 29 patients there was no fluorescence and quadrant biopsies were normal in all but three with moderate dysplasia. Within the minimum follow-up patients with bladder dysplasia detected by ALA-induced fluorescence had a higher risk of recurrence.


ALA-induced fluorescence of the bladder significantly enhanced the detection of dysplasia and in situ carcinoma. However, this technique requires further investigation using well-characterized instrumentation and study protocols to determine any effect on treatment choice.


In most cases bladder tumours diagnosed by ambulatory cystoscopy are not invasive lesions and have a low potential for progression, but a high recurrence rate [1,2]. To date, prognostic variables include the grade, number of tumours, location and size, which are all useful in daily practice for surveillance or to determine the need for adjuvant therapy. The association with flat urothelial dysplasia or carcinoma in situ (CIS) is an additional prognostic variable [3]. However, the detection of these lesions may be underestimated because they are heterogeneous and appear as normal bladder epithelium. Random biopsies in these cases are probably insufficient [4] and there are no available methods to locate and assess the extent of these neoplastic changes with an acceptable sensitivity.

In the last few years fluorescence techniques using 5-aminolaevulinic acid (ALA, a substrate of heme biosynthesis) have been developed to improve the detection of neoplastic lesions invisible with white light, and applied to oral, oesophageal, bronchial and urothelial epithelia [5–7]. For bladder application this technique relies on the intravesical instillation of ALA, with preclinical studies in models of normal and tumour-bearing rat bladder [8]. Intravesical ALA is converted into protoporphyrin IX by urothelial-cell cytoplasm. The associated enzyme malfunctions in tumour and dysplastic cells, resulting in the accumulation of protoporphyrin IX, and when the tumour or dysplastic cells are exposed to blue light, protoporphyrin IX fluoresces [9]. The pharmacology of ALA has been described; ALA is a stable component and is poorly absorbed into the systemic circulation [10]. Early reports showed that ALA-induced fluorescence improved the detection of urothelial lesions by 20% (from 73 to 97%) [11–14]. All these studies suggested that fluorescence cystoscopy is superior to white-light cystoscopy for detecting bladder tumours. In a previous study, d’Hallewin et al. [15] found that ALA-induced fluorescence was a very sensitive technique to detect bladder CIS and dysplasia. Herein, we report our experience of using ALA fluorescence to detect flat carcinomas of the bladder associated with a primary bladder tumour.


ALA fluorescence was used to detect lesions in 50 consecutive patients who had a bladder tumour and who were candidates for transurethral resection (TUR). The evaluation before surgery included urinary cytology and IVU; patients with a palpable T3 lesion were excluded from the study.

At 1–2 h before TUR, 50 mL of a solution of 8.4% sodium bicarbonate containing 1.5 g of ALA (pH 7) was instilled into the bladder through a disposable transurethral catheter. The solution was prepared previously and passed through a 20-µm filter to eliminate pyrogen agents. The exposure time was always 1–1.5 h.

Cystoscopy was then conducted using a high-intensity source of light (Storz, Tuttlingen, Germany) with a filter for blue light (380–450 nm) that induces a red fluorescence in positive cells. A 30° optical lens specially designed by the manufacturer with a yellow filter (Storz) was used to reduce the blue background and to improve the contrast of the fluorescent zones.

Cystoscopy was started using normal light, with a careful examination of the bladder, giving special attention to detecting abnormal zones. The characteristics of the tumour were also noted, particularly the size and number of lesions to treat. Then the bladder was examined again using blue light and all fluorescent areas biopsied. Using normal light, random biopsies were taken in the four quadrants of the bladder, and finally the tumour resected and sent for pathological assessment.


There were no side-effects related to ALA instillation; 45 patients had a primary bladder tumour and five an inflamed urothelium, with positive urinary cytology suggesting severe dysplasia. Twelve patients had a multifocal tumour. All bladder tumours were completely resected; the pathological details are shown in Table 1. Briefly, there were 35 TCCs and 15 inflammatory lesions (four inflammatory, nine nonspecific and two glandular cystitis).

Table 1.  The pathological grade and stage of urothelial carcinomas treated by TUR. All G2 and G3 tumours had positive cytology
StageGrade 1Grade 2Grade 3Total
pTa42  410
pT1a06  410
pT1b01  5  6
pT200  4  4
CIS00  5  5

In all cases the visible lesions (tumours and inflammatory) were strongly positive, showing intense fluorescence (Fig. 1). The bladder neck and prostatic urethra were always fluorescent but were less intense. The urothelium surrounding the tumour was slightly positive in benign and malignant cases. Distal from the tumour, fluorescent zones were considered as positive if the border was sharp and the surface homogeneous. In all, 21 patients had biopsies of a fluorescent zone, distal to the primary tumour. The pathological report showed four cases of grade 2 dysplasia, seven of grade 3 dysplasia and six of CIS; in four it was a nonspecific inflammatory reaction.

Figure 1.

a, Urothelial tumour viewed using conventional light cystoscopy and b, using blue light and after ALA instillation; the tumour appears fluorescent. c, An apparently normal bladder mucosae using light cystoscopy and d, the same zone, fluorescent using blue light.

All five cases of apparent dysplasia in conventional light were strongly fluorescent and biopsies showed CIS; Table 2 gives details of the biopsies in the fluorescent zones. Overall, ALA fluorescence detected bladder dysplasia at a significantly higher rate than did random biopsies (chi square, P < 0.01). This was also the case when ALA fluorescence was compared with urinary cytology, which was positive in all cases of TCC but did not predict the presence of dysplasia or CIS. All patients with a bladder tumour and associated dysplasia or CIS had BCG therapy, with two patients having a recurrence at 13 and 15 months.

Table 2.  The positivity of fluorescence of the bladder mucosa around the tumour and the histological result from biopsies, in parentheses, taken from positive zones or from random biopsies if negative
Positive fluorescence
Peritumoral  04 (0)5 (2)3 (1)4 (2)16 (5)
Flat lesions  00 (0)2 (1)3 (3)0  5 (4)
Negative florescence
Peritumoral  9200314
Flat lesions1515

Twenty-nine patients had no fluorescent zones in the urothelium surrounding the tumour. Histological analysis of random biopsies showed no abnormality in all but three patients, including one with inflammatory nonproliferating lesions and two with moderate dysplasia. After a minimum follow-up of 2 years none of these patients had a recurrence or progression.


Intravesical ALA fluorescence is reliable and reproducible in detecting bladder dysplasia. Despite the relatively few patients the results were significant when ALA fluorescence was compared with random biopsies. Although none of the patients with negative results had recurrence at 2 years the present data are insufficient to establish any correlation with recurrence rate and risk of progression.

The indications for ALA bladder fluoroscopy should be assessed according to the prognostic groups of superficial bladder tumours. Most superficial bladder tumours are low-grade TCC (G1–G2) and noninvasive Ta lesions, and rarely associated with dysplasia surrounding the tumour. In most cases these lesions will not progress. On the contrary, in undifferentiated (G2–G3) or invasive lesions (T1a, T1b or higher stages) the risk of progression is significantly higher, and it is important to know whether the tumour is associated with surrounding dysplasia. There might be a direct effect of the diagnosis of associated dysplasia on the need for adjuvant therapy; although it is a controversial issue for isolated bladder dysplasia [16], it is well accepted for CIS or dysplasia associated with an identifiable tumour [17].

From the present results and others [18,11] ALA fluoroscopy seems to better detect dysplasia than does urinary cytology or random biopsies. Detecting flat carcinoma may be particularly useful for endoscopic follow-up after BCG therapy. In a recent report using a similar technique [15] with Hypericin, another fluorescent pigment, the detection of flat carcinoma after BCG therapy was increased. For ALA-induced fluorescence, the accumulation of the compound depends on the cycling activity of the cells. Inflammatory lesions are positive in most cases, but in the context of active inflammatory lesions with symptoms. There is interest in using ALA-induced fluorescence to detect the recurrence of dysplasia in asymptomatic patients a few weeks after therapy.

Care is needed in interpreting positive fluorescent areas; the manufacturer recommends that a positive zone is ‘strongly’ positive when the borders are sharp and well identifiable. It may also be necessary to decrease the contact time between instillation and cystoscopy to distinguish inflammation (low proliferative activity) from dysplasia (high activity), although it may be difficult to apply a strict exposure time in clinical practice, given the variability of bladder capacity.

To date, studies on the use of ALA fluorescence in bladder cancer have focused on improving the detection of small lesions and the efficacy of resecting small tumours which are a factor in early recurrence [11,19]. In the present study all lesions were strongly fluorescent and ALA probably helped to detect satellite lesions or those in difficult areas. The detection of small lesions is particularly interesting in high-grade TCC, where the TUR should be as complete as possible. For this indication, Kriegmair et al.[20], in 104 patients, showed that ALA fluorocystoscopy improved the detection of lesions that were difficult to identify (small, multifocal, before resection, after BCG therapy). Compared with white-light cystoscopy the sensitivity of detection was improved from 73% to 97% using fluorocystoscopy. However, it appears that most inflammatory lesions are positive, which indicates that the technique has low specificity, although dysplasia may not be considered sufficient to modify treatment, especially moderate dysplasia. In the present study the risk of recurrence was higher in patients with dysplasia in fluorocystoscopy-based biopsies; this should be confirmed in a larger sample.

Although the present study included only 50 patients, and was thus insufficient to provide clear conclusions, the results suggest that fluorocystoscopy may provide useful information, improving the selection of bladder biopsies in positive zones and by helping the clinician to identify small or difficult lesions, providing a better idea of the immediate result of resection at the time of surgery. This technique is well adapted and has no side-effects. The results should be evaluated in a larger study, especially the effect of detected dysplasia on recurrence rate and the efficacy of adjuvant treatment.


Supported in part by a grant from ‘La ligue du Rhone’.


5-aminolaevulinic acid


transurethral resection


carcinoma in situ.