- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The improving knowledge of the surgical anatomy of the dorsal venous complex, the male external urethral sphincter, and the cavernosal nerves has substantially reduced the short- and long-term complications related to retropubic radical prostatectomy (RRP), allowing the widespread diffusion of this surgery . Urinary incontinence and erectile dysfunction (ED) remain the main long-term issues which can affect the quality of life (QoL) of operated patients.
Health-related QoL (HRQoL) consists not only of the patient's physical and psychological well-being, but also of their ability to work, live and act in a normal social setting. HRQoL can be perceived differently by patients of different ages, race, education, religion or from different social classes. Moreover, in oncological patients just the life-threatening nature of disease can compromise HRQoL .
Although there are several studies on urinary continence and sexual recovery after RRP, few authors report HRQoL after surgery. Moreover, most of the studies are biased by the absence of a baseline assessment of HRQoL, as they are cross-sectional rather than longitudinal studies . Thus the purpose of the present study was to assess prospectively the HRQoL in patients treated by RRP for clinically localized prostate cancer.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
From February 2002 to September 2003 all patients undergoing RRP in our department were invited to participate in this study. All patients were given a pack of questionnaires for the HRQoL assessment, to be completed through confidential self-administration and returned before surgery, and at 3, 6 and 12 months afterward. All patients who did not return the follow-up questionnaires in time were reminded by telephone 2 weeks after each deadline. In all, 150 patients had RRP during the evaluation period; 105 (70%) of them decided to participate, while 45 (30%) refused. Thirty patients (28.5%) were excluded from the study because of their delay in returning the questionnaires. The responses of the remaining 75 patients (71.5%) were analysed. To evaluate generic aspects of HRQoL we used the Italian validated version of the RAND 36-Item Health Survey 1.0 (SF-36) [4,5]. The SF-36 consists of 36 multiple-choice questions measuring eight distinct dimensions of current HRQoL, i.e. physical function, role limitations due to physical health problems, bodily pain, general health perception, emotional well-being, role limitations due to emotional problems, social function and energy/fatigue. Scores for each scale range from 0 to 100, with higher scores indicating higher level of function or well-being. The SF-36 was widely tested and validated in Italian general populations .
To the best of our knowledge, there are no Italian versions of any validated questionnaires assessing urinary incontinence. Hence, urinary continence was assessed through an institutional self-completed questionnaire. Patients were defined as continent if they used no pads or if they used only one pad as a precaution during the day or night.
Erectile function (EF) was assessed using the Italian version of the International Index of Erectile Function , using the score of the EF domain (questions 1–5, 15). Scores of 1–10 indicate severe, 11–16 moderate and 17–25 mild ED . Nerve-sparing surgery was performed only in patients aged < 65 years, with good preoperative EF (EF domain score ≥ 17), a total PSA level of ≤ 10 ng/mL and a biopsy Gleason score of < 8.
The SF-36 domain scores were reported as the mean (sd); as in previous studies, a patient was considered to have returned to the baseline value when the follow-up score returned to within 10 points of that before treatment on a particular scale [3,7]. The SF-36 domain scores at 1 year were compared to those of the corresponding age- and sex-related national population . The Mann–Whitney U-test was used to compare means of two unrelated samples, the Wilcoxon signed-rank test to compare the means of two related variables, the Friedman test to compare the means of more than two related samples and the Pearson chi-squared to compare categorical variables. In all statistical analyses a two-sided P < 0.05 was considered to indicate statistical significance. All study data were collected in a database and analysed with commercial software.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The mean (sd, range) age of the patients was 64.4 (5.6, 51–72) years; all were Caucasians. Sixty-seven patients (89%) were married, five (7%) unmarried, one (1%) separated and two were (3%) widowers. Educationally, 45 patients (60%) had no secondary school diploma, while 23 (31%) had a diploma and seven (9%) a degree. Twenty (27%) and 55 (73%) patients were working full- and part-time, respectively. For comorbidities, nine patients (12%) had diabetes, 32 (43%) had cardiovascular diseases (mainly hypertension), three (4%) had respiratory diseases and eight (11%) gastrointestinal diseases. Thirty patients (40%) had no comorbidities, 38 (51%) had one disease and seven (9%) two. Thirty-eight patients (51%) did not smoke cigarettes, seven (9%) were smokers and 30 (40%) had a long-term or recent history of cigarette smoke exposure. All patients were continent before surgery. Thirty-two patients (43%) had severe, and four (5%) and 12 (16%) had moderate or mild ED, respectively; 27 (36%) had no ED before surgery. A bilateral nerve-sparing procedure was used only in 17 of these 27 patients, specifically those aged ≤ 65 years and with a PSA level of < 10 ng/mL.
Comparing the baseline mean values of each SF-36 single-scale score to those at 3, 6 and 12 months, there was a statistically significant difference in physical function, role limitations due to physical health problems, role limitations due to emotional problems, and energy/fatigue (Table 1), but there was no statistically significant difference between baseline and follow-up values for the other four scales of the questionnaire. Moreover, mean baseline values overlapped the 12-month values for all the SF-36 scales. As to the time for the QoL to recover, in the domains where there was a statistically significant trend the 3- and 6-month scores were significantly lower than those at baseline. Table 2 reports the percentage of the patients whose scores overlapped the baseline values at follow-up.
Table 1. The mean (sd) SF-36 score variation in the different scales during the follow-up
|SF-36 scale||Baseline||3-months||6-months||12-months||Friedman test P|
|PF||91.52 (16.51)|| 81.01 (20.56)||86.01 (15.94)||88.80 (19.94)||<0.001|
|P|| ||<0.001||<0.001|| 0.61|| |
|RP||82.21 (32.08)|| 43.00 (44.46)||58.18 (43.04)||79.10 (35.92)||<0.001|
|P|| ||<0.001||<0.001|| 0.47|| |
|BP||88.21 (19.14)|| 84.03 (18.25)||84.03 (18.25)||85.26 (22.71)||0.74|
|P|| || 0.15|| 0.44|| 0.30|| |
|GHP||67.20 (18.53)|| 63.24 (17.45)||62.39 (18.31)||66.40 (20.21)||0.48|
|P|| || 0.10|| 0.03|| 0.59|| |
|EWB||66.12 (20.12)|| 67.40 (18.75)||67.16 (18.85)||72.24 (19.99)||0.28|
|P|| || 0.25|| 0.17|| 0.06|| |
|RE||71.90 (39.55)|| 50.90 (43.38)||63.40 (41.74)||77.21 (36.87)||0.001|
|P|| || 0.04|| 0.32|| 0.37|| |
|EF||69.65 (13.79)|| 60.57 (19.29)||61.79 (17.31)||68.78 (20.80)||0.01|
|P|| || 0.004|| 0.006|| 0.88|| |
|SF||79.82 (21.92)|| 73.60 (19.11)||75.09 (18.57)||81.39 (20.33)||0.22|
|P|| || 0.04|| 0.34|| 0.55|| |
Table 2. The percentage of patients reporting SF-36 scores which overlapped baseline values
|SF-36 scales||n/N (%) patients at |
|3 months||6 months||12 months|
|PF||39/75 (52)||58/75 (77)||67/75 (89)|
|RP||27/75 (36)||44/75 (59)||61/75 (81)|
|BP||49/75 (65)||57/75 (76)||64/75 (85)|
|GHP||53/75 (71)||57/75 (76)||63/75 (84)|
|EWB||61/75 (81)||71/75 (95)||73/75 (97)|
|RE||42/75 (56)||58/75 (77)||65/75 (87)|
|EF||46/75 (61)||57/75 (76)||61/75 (81)|
|SF||49/75 (65)||56/75 (75)||61/75 (81)|
The 12-month SF-36 data were similar to those of the healthy age-matched population (data not shown). At 1 year after RRP, an age >65 years, educational level less than secondary school, and pathological extracapsular extension of the primary tumour were associated with lower mean scores of some domains of the SF-36 (Table 3).
Table 3. Stratification of the mean (sd) SF-36 scores at 12 months after surgery according to clinical and pathological features
| married (67)||89.5 (19.3)||80.7 (34.3)||85.4 (22.6)||67 (19.7)||72.6 (20)|| 79.4 (35.3)||69.7 (20.2)||82.1 (20)|
| not married (8)||82.5 (24.6)||65.6 (48)||83.4 (24.9)||61.2 (24.7)||69 (20.1)|| 58.2 (46.3)||60.6 (24.7)||75.4 (22.9)|
| low (45)||84.7 (22.8)*||71.8 (40.2)*||80.4 (25.8)*||62.6 (20.1)||70.6 (20.8)|| 76.9 (37.5)||66.8 (21.1)||81.5 (21.1)|
| intermediate/high (30)||94.8 (12.5)||90 (25)||92.4 (14.5)||72 (19.2)||74.6 (18.7)|| 77.6 (36.5)||71.6 (20.3)||81.1 (19.3)|
| full time (20)||93.2 (12.8)||82.5 (32.5)||85.7 (18.8)||73.5 (16.2)||74.6 (17.3)|| 83.2 (31.6)||73.8 (20.1)||86.4 (14.9)|
| part-time (55)||87.1 (21.8)||77.8 (37.2)||85 (24.1)||63.8 (21)||71.3 (20.9)|| 75 (38.6)||66.9 (20.9)||79.5 (21.7)|
| absent (30)||89.1 (20.2)||78.6 (34.8)||88.7 (17.7)||73 (16.3)||72.8 (15)|| 75.4 (36)||70.9 (15.9)||85.5 (14.4)|
| present (45)||88.5 (19.9)||79.4 (37)||82.9 (25.4)||62 (21.4)||71.8 (22.8)|| 78.4 (37.7)||67.3 (23.5)||78.6 (23.2)|
|pT satge (TNM, 2002)|
| pT2 (44)||93.7 (11.5)*||86.9 (28.2)*||87.8 (19.9)||68.4 (18.9)||77.4 (17.8)*|| 85.4 (28.3)*||73.6 (19.1)*||86 (17.9)*|
| >pT2 (31)||81.7 (26.5)||68 (42.6)||81.5 (26)||63.5 (21.9)||64.8 (20.7)|| 65.4 (44.2)||61.9 (21.4)||74.8 (21.9)|
| present (69)||89.9 (18.5)||81.8 (33.4)||86.5 (21.3)||67.4 (19.5)||72.5 (19.9)|| 79.9 (34.4)||68.5 (20.9)||82.2 (19.8)|
| absent (6)||79.3 (29.2)||56.2 (49.5)||74.8 (32.1)||57.5 (24.6)||69.2 (21.2)|| 54.1 (50.2)||70.6 (21.1)||74.5 (24.1)|
| present (11)||98.6 (2.3)||95.4 (15)||95.4 (8.4)||79.5 (16.8)*||88.3 (10)*||100*||90 (12.2)*||92.4 (9.3)|
| absent (64)||87.1 (21.1)||76.2 (37.7)||83.5 (23.9)||64.1 (19.9)||69.4 (20)|| 73.2 (38.6)||65.1 (19.8)||79.4 (21.1)|
At 3 months after surgery 15 patients (20%) were incontinent, as were nine (12%) after 6 months and six (8%) after 12 months. Among the last group, four used one pad/day and two used two or more pads. The mean SF-36 scores reported by incontinent patients at 3, 6 and 12 months were lower than those reported by continent patients, although these trends were not statistically significant.
Of the 17 patients who had had a nerve-sparing RRP, 11 (65%) reported a 12-month EF domain score of >17. Patients with no ED had a significantly higher score for general health perceptions, emotional well-being, role limitations due to physical health problems, and energy/fatigue.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The lowest HRQoL of patients were at 3 months after RRP; during the follow-up there was a gradual improvement in the HRQoL scores in all the SF-36 domains, overlapping baseline values within a year in 80–90% of patients. For the domains where there was a statistically significant trend during the study (physical function, role limitation due to physical health problems, role limitation due to emotional health problems and energy/fatigue), the 3- and 6-month scores were significantly lower than those at baseline. Hence, we conclude that QoL recovers at 6–12 months, and that these statistically significant changes were in all cases clinically meaningful, because the changes in the scores were >0.5 sd of the baseline values, according to Guyatt et al..
Being >65 years old, having a lower educational level, pathological extracapsular extension of the tumour and ED was still associated with a worse QoL in the present patients at 12 months after surgery. On the contrary, patients with urinary incontinence had lower mean SF-36 scores, although these trends were not statistically significant.
Previous reports contain few data on the HRQoL of patients after RRP, with most results derived from cross-sectional studies with no baseline assessment [8–10]. In 1999 Litwin et al. published preliminary data from a longitudinal study using the self-administered SF-36 before and 3, 6 and 12 months after RRP. The final results of that study, published in 2001, showed that 3 months after surgery the baseline levels were recovered in 60% of the patients. Nine months after surgery 90% of the patients reached the baseline values, and the values remained stable up to 36 months after surgery . In 2001 Madalinska et al. reported similar data in a prospective longitudinal study using the SF-36 questionnaire, confirming the return to baseline HRQoL levels 12 months after surgery, and underscoring the minimal impact of RRP on psychological, physical and social aspects. Moreover, at 1 year after RRP the SF-36 scores were significantly higher than those of the age-matched healthy population. Similarly, Namiki et al. reported a longitudinal study of 72 Japanese patients, highlighting that 12 months after RRP the mean SF-36 scores were equal to or even better than pretreatment values. In that study, the baseline scores were recovered by 6 months after surgery.
For predicting the effect of RRP on HRQoL, Hu et al. suggested that age <65 years and the absence of comorbidity were variables which predicted the recovery of baseline levels of HRQoL. On the contrary, there was no relationship between HRQoL and clinicopathological tumour features. From the present data, patients aged >65 years had lower mean scores for physical function, role performance with physical limitations, mental health, role performance with emotional limitations, and energy/fatigue status. However, all the scores recorded in both age groups were similar to or better than those of the reference national healthy population .
In the present study, patients with pathological extraprostatic disease had significantly lower mean scores than those with localized tumours in six of the eight scales (physical function, role performance with physical limitations, role performance with emotional limitations, social function, mental health, and energy and fatigue status). For the impact of prostate cancer prognostic factors on HRQoL, Litwin et al., analysing 452 patients undergoing watchful waiting, surgical treatment or radiotherapy for prostate cancer, underscored that needing adjuvant treatments, or the possibility of biochemical or clinical relapse of the disease, could significantly impair the QoL scores in such patients. These data should be verified in further prospective studies with more patients. Nevertheless, it could be important information to consider when discussing tumour stage, therapeutic options and prognosis with the patient.
Urinary incontinence and ED can worsen the HRQoL of patients after RRP . The reported prevalence of these complications is very variable, mostly according to the different definitions and instruments used to assess them. Moreover, patient features before treatment, surgical technique and the surgeon's experience could increase such variability. The prevalence of urinary incontinence after RRP progressively reduces during the follow-up; at 1 year after treatment, according to the definition of ‘need for protection’, the percentage of patients with urinary incontinence is 1–35%[12,17]. In the present study the mean SF-36 scores of patients with urinary incontinence after surgery were lower than those of continent patients, although the trend was not statistically significant. Possibly this was because there were few incontinent patients and they had only mild incontinence. Relevant data were provided by the Prostate Cancer Outcome Study (PCOS), a large prospective, population-based cohort study involving >2000 patients . That study analysed patients undergoing different prostate cancer treatments (RP, radiation therapy, hormonal ablation therapy, or watchful waiting), showing a relevant effect of urinary incontinence on five SF-36 domains at 2 years after the primary treatment.
For ED, bilateral nerve-sparing RRP allows the recovery of EF in 56–86% of patients [19,20]. Preserving the neurovascular bundles is mostly indicated in younger patients (≤65 years old), with good preoperative EF, no comorbidities and with a high probability of having organ-confined disease (a PSA level of ≤ 10 ng/mL and/or a Gleason score of ≤ 7) . In the present study only 23% of patients had such features; in this subgroup at 1 year after surgery, potency rates were similar to those reported elsewhere. The reported impact of ED on HRQoL is controversial; according to Kao et al., ED was an independent predictor of lower HRQoL, and was one of the main reasons not to repeat RRP in the theoretical case of need. Using the Prostate Cancer Index, Litwin et al. reported the recovery of sexual function a 1 year after surgery in only a third of patients; in the same study, at 1 year after surgery 90% of patients had HRQoL scores equal to that at baseline. On the contrary, Gralnek et al. reported higher scores for physical role and physical limitation domains of the SF-36 in patients undergoing nerve-sparing surgery than in those who did not have a nerve-sparing procedure. In the PCOS study, Penson et al. showed an influence of ED on general HRQoL, with impairments of the scores of six SF-36 domains. Moreover, ED-related dissatisfaction can be influenced by the response to pharmacological therapy. Perez et al. found that pharmacological therapy, mechanical penile prostheses or vacuum devices might allow slightly better functional results than nerve-sparing RRP alone. In the present study the patients who recovered EF with or without using phosphodiesterase 5 inhibitors had higher SF-36 scores than those who did not, in the domains of general health, psychological well-being, emotional role, and energy/fatigue status. Considering the strict criteria adopted in selecting candidates for nerve-sparing surgery, those data could be more easily explained by the pretreatment features of these patients than by functional results.
The present study has some limitations. First, only half of the patients treated by RRP completed the study because some refused to participate or did not return the questionnaires. Second, we did not use a disease-specific questionnaire to assess the QoL of the patients. There are some questionnaires specific for patients with prostate cancer, e.g. the Prostate Cancer Index or the Expanded Prostate Cancer Index Composite, but to our knowledge there are no Italian-validated versions of such questionnaires. Similarly, we had to use an institutionally developed questionnaire to evaluate urinary continence. Moreover, the size of the present cohort might have impaired some of the statistical analyses, which might be under-powered.
In conclusion, assessing HRQoL is an important endpoint for evaluating the results of radical treatment for prostate cancer. Advances in surgical treatment allow high urinary continence rates and satisfactory potency rates after nerve-sparing surgery. There was a statistically significant trend during the study in some domains of the SF-36, with the 3- and 6-month scores being significantly lower than those at baseline. Moreover, the data suggest that QoL recovers in 6–12 months. Age, the educational level of patients and local extension of the tumour could significantly affect the HRQoL scores. Moreover, patients potent after RRP had better QoL scores than those who had ED.