Does Transurethral Resection of the Prostate Pose a Risk to Life? 22-Year Outcome

0022-5347/95/1535-1506$03.00/0

Vol. 153,1506-1509,May 1995 Printed in U.S.A.

THE JOURNAL OF UROLOGY

Copyright 0 1995 by A M E m c m U R O ~ I CASSOCIATION, AL INC.

DOES TRANSURETHRAL RESECTION OF THE PROSTATE POSE A RISK TO LIFE? 22-YEAR OUTCOME KEN KOSHIBA, SHIN EGAWA, MAKOTO OHORI, TOYOAKI UCHIDA, EIJI YOKOYAMA KIYOSHI SHOJI

AND

From the Department of Urology, Kitasato University School of Medicine, Kanagawa, Japan

ABSTRACT

The clinical outcomes of 717 patients who underwent transurethral resection for benign prostatic hyperplasia between 1971 and 1981, and of 48 who underwent open prostatectomy during the same period were evaluated. All living patients could be followed for a minimum of 12 years postoperatively. The cumulative percentage of patients undergoing a secondary operation was substantially greater after transurethral resection of the prostate than after open prostatectomy. The volume of resected tissue, operating time, requirement for blood transfusion and hyponatremia during or after the procedure did not affect long-term outcome of patients in the transurethral resection group. Abnormal preoperative electrocardiography and azotemia appeared to be associated with increased risk of postoperative mortality after controlling for other variables (p c0.05).Actuarial survival rates did not differ substantially for patients who underwent transurethral resection and open prostatectomy, and exceeded the expected survival rates in the general male population in the same age group in Japan. Both procedures are safe for the symptomatic relief of urinary obstruction due to benign prostatic hyperplasia. Prostatectomy does not jeopardize long-term survival of the patients. Kn! WORDS:prostate, prostatectomy, survival

Reports from Manitoba, Canada and Maine have attracted the keen attention of urologi~ts.l-~ The safety and efficacy of prostatectomy, particularly for transurethral resection of the prostate gland, have come into question. Insurance claim data indicate patients to have higher rates of cystoscopy, reoperation and urethral stricture following transurethral resection of the prostate. More significantly, transurethral resection patients have higher mortality rates than those treated by open prostatectomy. Transurethral resection of the prostate would, thus, appear to decrease life expectancy.1" However, there is no known biological explanation as to why transurethral resection of the prostate would lead to increased mortality years after surgery. The choice for transurethral resection or open prostatectomy is based on many considerations in cases when benign prostatic hyperplasia (BPH) requires operation. Information on patient comorbidity is often limited and, consequently, the basis for selection of operation cannot be determined with certainty. Concato et al considered inaccurate assessment of the severity of illness in previous studies to be possibly the reason for the higher mortality rates found in patients following transurethral resection of the prostate.6 Administrative data were insufficient for the analysis of patient survival. Because of the small sample size, however, their conclusion required additional confirmation. Approximately 4,000 prostatectomies have been conducted at our institution during the last 22 years, which has made it possible for us to study the long-term outcomes of prostatectomy. Extensive followup studies were conducted on patients who underwent transurethral resection or open prostatectomy. PATIENTS AND METHODS

Between August 1971 and December 1981, 836 patients underwent prostatectomy for BPH at our university hospital. In 71 of these cases (8.5%)cancer was incidentally noted in the resected specimens and such patients were excluded from further study. Survival rates and causes of death as of FebAccepted for publication September 16, 1994

ruary 1993 were tabulated for the remaining 765 patients who underwent transurethral resection of the prostate (717) or open retropubic prostatectomy (48). Effects on long-term survival were evaluated for resected volume of the prostatic adenoma, operating time, requirement for blood transfusion, development of hyponatremia during or after the procedure, preoperative abnormal electrocardiograms indicating ischemic heart change or arrhythmia and azotemia. Clinical characteristics, and morbidity and mortality postoperatively were compared for patients undergoing transurethral resection of the prostate and open prostatectomy. Causes of death were confirmed from hospital records with death certificates. For patients who were lost to followup, the last date of visit to the clinic postoperatively was used to calculate interval under observation. Survival curves were constructed by the Kaplan-Meier method.7 Expected survival rates in the general male population were determined from Japanese population tables by age (&year age groups) and calendar year.8 Statistical significance was determined using Student's t test (2-tail, nonpaired) or log-rank test with p <0.05 as significant. The prognostic value of the different variables determined a t operation was further analyzed using Cox's proportional hazThe ards regression model with p c0.05 being ~ignificant.~ analyses were performed with a statistics software package. RESULTS

The mean age of the 765 patients who underwent prostatectomy was 69.8 years (range 47 to 96 years). Surgery was performed with the patient under epidural (95%), general (4%)or spinal (1%)anesthesia. Vasectomy was done in 64.2% of the patients. All living patients could be followed for a minimum of 12 years (mean 15.1, range 12 to 22). Mean age of the patients who underwent transurethral resection or open prostatectomy was 69.7 and 69.9 years, respectively. No difference was found between both groups (table 1). Average operating time was much longer in patients who had undergone open prostatectomy (108.5 minutes) than transurethral resection of the prostate (68.8 minutes). The mean volume of

1506

1507

LONG-TERM SURVIVAL OF PATIENTS FOLLOWING PROSTATECTOMY TABLE1. Comparison of clinical characteristics of transurethral resection and open prostatectomy patients

TABLE3. Causes of death following prostatectomy No.

Mean (range) Transurethral Resection of Open prostate Prostakctomy (717 DtS.) (48 pts') Age (yrs.) Operating time (mins.) Wt. resected tissue (gm.) Repeat resection (%I

Transurethral Statistical Significance ( p value) Cardiac Malignancy Cerebrovascular brig Pneumonia Chronic renal failure Accident Miscellaneous Unknown Totals

69.7 (47-96) 69.9 (52-84) Not significant 68.8 (10-175) 108.5 (45-275) <0.001 31.1 (2-177) 53.9 (19-130) <0.001 5.6 (40/717) 2.1 (V48) <0.05

resected tissue was significantly greater in the open prostatectomy (53.9 gm.) than in the transurethral resection (31.1 gm.) group. The largest adenoma (177 gm.) was removed transurethrally during a 1-stage procedure. Of the 717 patients in the transurethral resection group 40 (5.6%)underwent secondary resection because of postoperative contracture of the bladder neck or inadequate initial resection of the adenoma, which subsequently led to initial symptoms. Of the 48 patients in the open prostatectomy group 1 (2.1%)underwent transurethral resection for recurrent BPH that caused urinary obstruction (table 1). Among the 7 17 transurethral resection patients 254 were alive and 318 were dead at the time of this study, while 145 were lost to followup. Of the 48 open prostatectomy patients 20 were alive, 22 were dead and 6 were lost to followup (table 2). Operative mortality rate (defined as death within 1 month postoperatively) was zero in this series. Two patients in the transurethral resection group died of pneumonia and a cerebrovascular accident, respectively, within 3 months postoperatively. No patient in the open prostatectomy group died within 3 months. Causes of death were confirmed for 329 of the 340 patients (96.8%). The most common causes were cardiac disease, malignant tumor, cerebrovascular accidents and aging. No patient died of prostate cancer. Thus, there appeared to be no differences in the causes of death for the 2 methods (table 3). The overall survival probability for the patients did not differ significantly between the transurethral resection and open prostatectomy groups (p >0.05, fig. 1). Expected survival rates from the same age group in the Japanese general male population remained consistently lower throughout the observation period, especially within the first 20 years postoperatively. No increased risk of death in prostatectomy patients could be detected. Figure 2, a presents actuarial survival curves of patients who underwent transurethral resection of the prostate, plotted against resected adenoma weight. Resected tissue weighed less than 20 gm. in 38.1%of the patients, 20 to 60 gm. in 49.8% and more than 60 gm. in 12.1%. Operating time was less than 60 minutes in 45.9%of the patients, 60 to 90 minutes in 36.7%and longer than 90 minutes in 17.4%.Hyponatremia of serum sodium less than 125 mEq./l., was observed in 10.9% of the patients who underwent transurethral resection of the prostate. Blood transfusion was required by 32.4%of the patients during or after the operation (fig. TABLE2. Survival and death of patients at 12 to 22 years after Drostatectomv No. (%I Transurethral

Resection of Prostate Alive Dead Unknown Totals

R

254 (35.4) 318 (44.4) 145 (20.2) 717 (100)

Open Prostatectomy

Totals

20 (41.7) 22 (45.8) 6 (12.5) 48 (100)

274 (35.8) 340 (44.4) 151 (19.8) 765 (100)

~

prostatectomy ~ ~ ~ o t a ~~ s

(717 pts.)

(48 pts')

83 (26.1) 71 (22.3) 43 (13.5) 43 (13.5) 22 (6.9) 7 (2.2) 6 (1.9) 33 (10.4) 10 (3.2) 318 (100)

6 (27.3) 2 (9.1) 3 (13.6) 4 (18.2) 3(13.6) 0 (0.0) 0 (0.0) 3 (13.6) 1 (4.6) 22 (100)

~

89 (26.2) 73 (21.5) 46 (13.5) 47 (13.8) 25 (7.4) 7 (2.0) 6 (1.8) 36 (10.6) 11 (3.2) 340 (100)

- TURP

1 W

(n=717) --(na) -OpenP. Exp. .urvival rate

s

h

v

2

(%)

Open

75-

.-I $ * -m L

v)

.-

TURP

a

Open

O-

1

No censored pts

3 255 I

0

US.

5yr l O y r 15yr 2Oyr 25yr I25 137 289 387 399 6 6 11 25 25

b

A I

$0

15

do

I

25

Years after prostatectomy

FIG. 1. Actuarial survival rates for transurethral resection (TURF') and open prostatectomy ( P )groups, compared with expected (Exp.)survival rates of general male population of same age.

2, b , c and d). Abnormal electrocardiogram results were observed in 31.9%of the patients. Azotemia was noted in 4.7% of the transurethral resection patients (fig. 2, e and f ) . These results were further analyzed using the Cox proportional hazards regression model (table 4). Age a t operation, presence of azotemia and abnormal electrocardiogram were all significantly associated with increased risk of postoperative mortality in the univariate and multivariate analyses (p <0.05). The relative risk for postoperative mortality was 1.65 in patients with elevated levels of serum creatinine of 1.6 mg./dl. or greater compared to those with normal levels. Cardiovascular accidents were the most common cause of death in this group, accounting for 40%of the 25 cases. b n a l failure was responsible only for 3 deaths. The relative risk for mortality was 1.28 in patients with abnormal electrocardiography findings compared to those who had a normal result. Cardiovascular accidents were also the most common cause of death in this group, accounting for 50% of the 114 cases. Interestingly, there was a significant survival advantage in patients with larger glands resected. The cause of this fact is not readily certain, however, it appeared that resected adenoma weight did not jeopardize patient survival. DISCUSSION

In Japan as well as in most western countries, transurethral resection of the prostate has gradually replaced open prostatectomy as the choice of surgery for BPH. Indication for an open procedure is usually limited to extraordinarily large glands. Based on insurance claim data, Roos and Ramsey presented evidence for increased rates of reoperation and decreased survival after transurethral resection of the prostate compared with open prostatectomy.2 These surprising

LONG-TERM SURVIVAL OF PATIENTS FOLLOWING PROSTATECTOMY

1508 a.

m.R.crNrn)

(m

-
--1(uD(-

0

s

tb

io

i5

C. loo-

b.

M(rn14)

-,im(mr)

E

.Va(0?q

0

S

Years after TURP

I

b

1

’

5

i

o

i

5

0-

b

5

Years after TURP

1

0

1

5

2

0

2

5

Years after TURP

FIG.2. Survival rates in transurethral resection group as function of volume of resected (Wt. resect.) tissue ( a ) , operating time ( b ) , development of hyponatremia ( c ) ,requirement of blood transfusion ( d ) ,abnormal electrocardiogram(ECG)findings ( e )and azotemia ( f ). s-Nu, serum sodium. s-Crea, serum creatinine. TABLE4. Prognostic value of clinical characteristics following prostatectomy Variables Age (yrs.): Less than 60 60 to 69 70 to 79 80 or more Serum creatinine (mg./dl.): Less than 1.6 1.6 or more Electrocardiogram: Normal Abnormal Hyponatremia (125 mEq./l. or less): Neg. Pos. Operating time (mind: Less than 60 60 to 89 90 or more Wt. resected tissue (gm.): Less than 20 20 to 59 60 or more Blood transfusion: No Yes

No. ( I ) 74 (10.3) 269 (37.5) 309 (43.1) 65 (9.1) 683 (95.3) 34 (4.7)

Univariate Analysis (confidence interval)

Multivariate Analysis (confidence interval)

1.00 2.38 (2.20-2.56)* 5.66 (4.867.46)t 13.47 (12.67-15.27)t

1.00 2.42 (2.22-2.62)t 5.85 (5.65-6.05)t 14.17 (13.97-14.37)t

1.00 2.17

(1.71-2.63)t

1.00 1.65

(1.29-2.01)$ (1.13-1.43)$

488 (68.1) 229 (31.9)

1.00 1.31

(1.15-1.45)$

1.00 1.28

639 (89.1) 78 (10.9)

1.00 0.79

(0.634.95)

1.00 0.90

(0.72-1.08)

329 (45.9) 263 (36.7) 125 (17.4)

1.00 0.97 1.04

(0.90-1.04) (0.97-1.11)

1.00 1.04 1.08

(0.9P1.14) (0.98-1.18)

273 (38.1) 357 (49.8) 87 (12.1)

1.00 0.93 0.86

(0.85-1.01) (0.78-0.94)

1.00 0.78 0.61

(0.69-0.87)’& (0.52-0.70)$

485 (67.6) 232 (32.4)

1.00 1.00

(0.8S1.12)

1.00 0.89

(0.77-1.01)

* p <0.05.

t p <0.0005. ’& p <0.03.

findings raised significant concern in the urological community. Although the inadequacy of data for assessing patient co-morbidity may partially explain this finding, other studies demonstrate increased relative risk of death after transurethral resection of the prostate even after controlling patient characteristic^.^.^ Whether transurethral resection of the prostate actually contributes to decreased survival rates is difficult to confirm, since such studies were not done with this objective specifically in mind. Recently, Concato et a1 conducted a n extensive review of the medical records of 252 patients who underwent transurethral resection or open prostatectomy for treatment of BPH.6Although crude 5-year mortality rates were higher in the transurethral resection group, the adjusted risk of transurethral resection of the prostate decreased after controlling for severity of co-morbidity. Problems in previous studies may have arisen from the

use of administrative rather than detailed clinical information, making accurate assessment of the seventy of co-morbidity impossible. Transurethral resection did not appear to increase long-term mortality. Because of small sample size, their conclusions require additional confirmation despite the sophisticated methods of analysis. Transurethral resection of the prostate gland has long been the only accepted therapeutic option for BPH a t our institution, especially during the last 12 years. The numbers of transurethral resections performed during the last 22 years (3,831 cases) contrasted vividly with the few numbers of open prostatectomies (48). Our experience in treating BPH by transurethral resection of the prostate or open prostatectomy during a long period differs considerably with that in Manitoba, Canada. None of our patients died within 1 month postoperatively. Only 2 patients died within 3 months after

LONG-TERM SURVIVAL OF PATIENTS FOLLOWING PROSTATECTOMY transurethral resection but none died i n the open prostatectomy group. These 2 deaths in the early postoperative days were unrelated to the surgical procedure. The long-term survival probability of patients did not differ significantly between t h e transurethral resection and open prostatectomy groups, and survival rates exceeded those expected for the general male population i n t h e same age group. There was a substantial difference in the proportion of patients who were lost to followupapproximately 20% i n the transurethral resection group and 12% i n the open prostatectomy group. The lack of difference between groups could be reversed if a significant proportion of the unknown group had actually been dead. However, the survival rates between groups did not differ significantly even in the extreme situation i n which all those lost to followup were assumed to be dead after the last visit ( p = 0.79). In our study, preoperative serum creatinine levels of 1.6 mg./dl. or greater and abnormal electrocardiography results were associated with a decreased chance of long-term survival. These were independent prognostic factors besides age at operation. Cardiovascular accidents were the most common cause of death i n these patients. Patients with these conditions preoperatively were reported to have significantly more postoperative complications.10 The few numbers of patients with azotemia in our study preclude drawing any definite conclusion. Geographic variationin mortality from transurethral resection of the prostate has been reported to be slight in western Since our population consisted invariably of oriental men, racial factors should also be considered when analyzing our data. Transurethral resection certainly is associated with a greater risk of reoperation. In general, patients with smaller glands undergo transurethral resection rather than the open procedure, which i n t u r n leads to more frequent postoperative contracture of the bladder neck. Urinary i m t a t i v e symptoms, particularly in patients with smaller glands, may occur through activation of the sympathetic nervous system and, thus, cannot be corrected by transurethral resection of the prostate alone. Endoscopic manipulation itself has potential risks of urethral stricture at a later date. Operative failure may be encountered more often i n such patients.l1,l2 More than twice t h e number of patients i n the transurethral resection group underwent secondary resection due to recurrent symptoms, compared with 2.1%in those who underwent open prostatectomy i n this study. These results a r e compatible with those at other institutions.'-5 The relatively high percentage of patients undergoing transurethral resection of the prostate who required blood transfusion is considered due to limitation i n t h e quality of the resectoscope used at the time. Visualization of t h e operative field was insufficient due to the unavailability of a fiberoptic and micro-lens system 20 years ago, often causing severe blood loss during and after the procedure. The incidence of patients requiring blood transfusion has decreased dramatically to 3.6% during the last 3 years.

1509

CONCLUSIONS

Many factors should be considered when choosing either transurethral resection of t h e prostate or open prostatectomy for the symptomatic relief of urinary obstruction. Our longterm experience is in accord with the conclusion of Concato e t a1.6 However, an adequate scientific basis based upon prospective randomized study is still needed to guide physicians and patients in making t h e choice of surgery. Cooperative clinical trials are to be conducted on a large scale under the direction of t h e American Urological Association. Such data should facilitate the evaluation of transurethral resection for the symptomatic relief of urinary obstruction due to BPH.

REFERENCES

1. Wennberg, J. E., Roos, N., Sola, A,, Schori, A. and JaEe, R.: Use of claims data systems to evaluate health care outcomes. Mor-

tality and reoperation following prostatedomy. J.A.M.A., 257: 933,1987. 2. Roos, N. P. and Ramsey, E. W.: A population based study of prostatectomy: outcomes associated with differing surgical approaches. J . Urol., 137: 1184,1987. 3. Roos, N. P.,Wennberg, J. E., Malenka, D. J., Fisher, E. S., McPherson, K, Andersen, T. F., Cohen, M. M. and Ramsey, E.: Mortality and reoperation after open and transurethral resection of the prostate for benign prostatic hyperplasia. New Engl. J . Med., 320 1120,1989. 4. Wennberg, J. E., Mulley, A. G., Jr., Hanley, D., Timothy, R. P., Fowler, F. J., Jr., Roos, N. P., Barry, M. J., McPherson, K , Greenberg, E. R., Soule, D., Bubolz, T., Fisher, E. and Malenka, D.: An assessment of prostatectomy for benign urinary tract obstruction. Geographic variations and the evaluation of medical care outcomes. J.A.M.A., 259 3027, 1988. 5. Malenka, D.J., Roos, N., Fisher, E. S., McLerran, D., Whaley, F. S., Barry, M. J., Bruskewitz, R. and Wennberg, J. E.: Further study of the increased mortality following transurethral prostatectomy: a chart-based analysis. J . Urol., 144.224,1990. 6. Concato, J., Horwitz, R. I., Feinstein, A. R., Elmore, J. G. and Schiff, S. F.: Problems of comorbidity in mortality after prostatectomy. J.A.M.A., 267: 1077,1992. 7. Kaplan, E. L. and Meier, P.: Nonparametric estimation from incomplete observations. J. h e r . Stat. Ass., 53: 457,1958. 8. Statistics and Information Department, Minister's Secretariat, Ministry of Health and Welfare. The 16th Life Tables, p. 176, 1987. 9. Cox, D.R.:Ftegression models and life tables. J. Roy. Stat. Soc. B, 34: 187, 1972. 10. Mebust, W. K,Holtgrewe, H. L., Cockett, A. T. K, Peters, P. C. and Writing Committee. Transurethral prostatectomy: immediate and postoperative complications.A cooperative study of 13 participating institutions evaluating 3,885 patients. J. Urol., 141: 243,1989. 11. Hohenfellner,R. and Jonas, U.: Results following prostatectomy. Urol. Int., 33: 187,1978. 12. Turner Warwick, R., Whiteside, C. G., Arnold, E. P., Bates, C. P., Worth, P. H. L., Milroy, E. G. J., Webster, J . R. and Weir, J.: A urodynamic view of prostatic obstruction and the results of prostatectomy. Brit. J . Urol., 45:631,1973.