© 1991 Karger AG. Basel 0302-2838/91/0206-0011 $ 2.75/0
Eur Urol 1991;20(suppl 2):11-17
Benign Prostatic Hyperplasia: A Population-Based Study 1605949
C.G. Chute*, W.P. Stephensonb, H.A. Guessh, M. Liebera JMayo Clinic, Rochester. Minnesota, and '’Merck Sharp & Dohme Research Laboratories. West Point. Pennsylvania. USA
Key Words. Prostatectomy • Incidence ■ Epidemiology, population-based Abstract. To evaluate the incidence and outcome of initial surgery for benign prostatic hyperplasia (BPH) and to clarify the natural occurrence and progression of such urologie diseases, two studies have been conducted in a freeliving population in Rochester, MN. The first followed 330 men who had not been diagnosed with prostate or bladder cancer at the time of prostatectomy. All surgery subjects were area residents and between 46 and 95 years of age (mean age 70 years). Among the operated subjects, 14 (4.2%) had serious intraoperative complications, 32 (9.7%) were rehospitalized for urologie complications within 30 days after surgery, and 13 (3.9% ) experienced other serious compli cations in that same time period. Blood transfusions within 30 days of surgery were necessary in 45 patients ( 14% ). The risk of reoperation within óyears of the initial surgery was calculated at 15.1% (95% Cl: 9.7,20.6). On the basis of ageand sex-specific mortality statistics for Rochester, short- and long-term postoperative mortality was not statistically sig nificantly different from that expected. Results of the second study are not yet available. This population-based evalua tion of the natural history of urologie disease is expected to clarify the relative utility of various treatment options and provide a useful perspective on the management of BPH.
tion of a nonreferred study group, and thus provide rel atively unbiased data that derive from a free-living popu lation. Lifetime inpatient and outpatient medical records from the surrounding health care providers were used for Study 1, independent of the patient’s referral to our clinic. The Rochester Epidemiology Project enabled us both to establish a well-defined population base for determining incidence and outcome and to limit our surgical study cohort to men undergoing prostatectomy for the first time and specifically for BPH. This same resource permitted the enumeration of Olmsted County residents around the city of Rochester, from which a random sample could be selected for our natural history cohort (Study 2).
Patients and Methods Study 1: Surgical Intervention for BPH Aprimarytool ofStudy 1 was the Rochester Epidemiology Project, a record-linkage system whereby all medical records from the past sev eral decades have been assembled for Rochester residents [11]. Patient
Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 8:26:52 AM
Benign prostatic hyperplasia (BPH), one of the most common conditions associated with aging in men, has been noted at autopsy in approximately 40% of men in their 50s and in up to 70% of those in their 60s [1]. Reflect ing the high incidence of BPH, an estimated 380,000 trans urethral prostatectomies (TURPs) were performed in the United States in 1987, 97% of which were in men over age 54 [2]. In recognition of the frequency of prostatic sur gery for BPH, a population-based study (Study 1) was conducted to evaluate the incidence and outcome follow ing initial prostatectomy for BPH in a general population, in this case residents of Rochester, MN. As an expansion to this postsurgery study, another population-based cohort (Study 2) from the same geographic area is cur rently being studied to clarify and describe the incidence and natural history of BPH and other urologie diseases in a free-living population. In contrast to most other studies of BPH, which tend to represent highly selected patient groups over a short term [3-10], these two studies are unique in their evalua
Chute/Stephenson/Guess/Lieber
12
Table 1. Characteristics of men having first surgery for benign prostatic hypertrophy in Rochester, MN (1980-1987) Characteristic Age 45-54 55-64 65-74 75-84 85 + Smoking History Current smoker Ex-smoker Never smoked Current nonsmoker (NOS) Unknown Alcohol Use None Occasional/regular Heavy drinker History of alcoholism Unknown Medical History Heart Disease CHD MI Coronary artery bypass/PTCA Congestive heart failure Hypertension Valvular heart disease Congenital heart defect Cerebral vascular disease Transient ischemic attack Stroke Peripheral artery disease IC Gangrene/amputation (vascular insufficiency) Peripheral artery bypass COPD Diabetes mellitus Cancer History of deep vein thrombosis History of pulmonary embolis Bleeding disease
No. Patients (%)
11 89 124 87 19
(3.3) (27.0) (37.6) (26.5) (5.8)
63 164 51 49 3
(19.0) (49.7) (15.5) (14.8) (0.9)
94 190 23 13 10
(28.5) (57.6) (7.0) (3.9) (3.0)
118 (35.8) 103 (31.2) 61 (18.5) 20
(6.0)
28 (8.5) 128 (38.3) 22 (6.7) 2
(0.6)
44 26 31 31 29
(13.3) (7.9) (9.4) (9.4) (8.8)
1 3 80 42 33 10 4
(0.3) (0.9) (24.2) (12.7) (10.0) (3.0) (1.2)
2
(0.6)
NOS = Not otherwise specified; CHD = coronary heart disease; MI = myocardial infarction; PTCA = percutaneous transluminal coro nary angioplasty; IC = intermittent claudication; COPD = chronic obstructive pulmonary disease. From Stephenson et al.12 with permission.
inpatients, on death certificates, and at autopsy. All surgical proce dures performed on Rochester residents at Mayo-affiliated hospitals and at Olmsted Community Hospital also are available in the file. Such a data-retrieval system provides a unique resource for populationbased study. The Study 1 population comprised all men who had undergone prostatectomy for BPH between January 1, 1980. and December 31, 1987. They had been Rochester residents for at least 1 year prior to sur gery, and had not been diagnosed with prostate cancer. Of the 389 men who satisfied the entry criteria, 13 (3.3%) were excluded as a result of histologic (n = 2) or clinical (n=l) evidence of prostate or bladder cancer prior to surgery. Another 46 (12%) were excluded due to prior prostatectomy. A more detailed description of our prostatectomy series (Study 1) methods appears elsewhere [12]. Briefly, collected data included demographics, medical history, urologie history, course and outcome of index surgery and hospitalization, length of hospital stay, subse quent urologie hospitalizations and visits, urologie symptoms within 3(1 days prior to or at any time following surgery (when documented in the medical records), and vital status. Patients’ records were carefully abstracted for evidence of serious intraoperative complications and for surgical mortality and serious surgical morbidity. Annual incidence was calculated as both crude and age-adjusted rates, based on the 1980 figures for the US white male population 45 years of age and over [13]. Changes in incidence with time were analyzed by Poisson regression [14]. Multiple regression analysis was used to evaluate the effects of age and year of surgery on length of hospital stay [15], and logarithmic transformation was used to adjust for the extreme skew of the distribu tions. Risks of reoperation, prostate cancer, and mortality following surgery were analyzed using life table methods (Kaplan-Meier prod uct-limit method) [ 16]. and exact confidence limits for the standardized mortality ratios (SMR) were calculated by Austin’s method [17]. Study 2. Natural History of BPH Study 2 used a population-based survey that randomly sampled 2,000 male residents of Olmsted County, MN, thus drawing from the same population as Study 1. All participants were between the ages of 40 and 80 years. Home visits were conducted and included measure ment of urine flow rates (Dantec 1000 Uroflow, Dantec Elektronik, Skovlunde. Denmark), completion of a validated quality of life and urologie symptom questionnaire, and physician assessment of family history, medication use, and living status [18]. Subsequently, a 500member random subset of these men underwent in-clinic evaluation of prostate size and status via digital rectal examination, standardized prostatic ultrasound, abdominal ultrasound measure of urinary reten tion. prostatic specific antigen, and serum creatinine. Repeat question naire and physical examinations were also conducted at the in-clinic visit. These metrics are expected to be repeated on the 2,000 member cohort 18 to 24 months following enrollment to establish longitudinal change.
records are easily retrievable via a computer-based master file and incorporate outpatient visits to virtually every medical care provider serving the local community (regardless of affiliation with the Mayo Clinic) and cover office consultations, emergency room visits, house calls, and nursing home visits, as well as diagnoses recorded for hospital
Demographic, behavioral, and clinical characteristics of the 330 men in Study 1 are listed in table 1. All of the subjects were at least 45 years of age and 97% were 55 years or older; all but one were white. The most common
Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 8:26:52 AM
Results
Benign Prostatic Hyperplasia
13
Table 2. Urologie complaints within 30 days prior to surgery for benign prostatic hypertrophy (Study 1)
Table 3. Age-specific incidence of surgery for BPH in Rochester, MN (1980-1987)
Symptom
No. patients (%)
Age
Nocturia Reduced force of stream Frequency Incomplete emptying Hesitancy Complete retention Urgency Dysuria Postvoid dribbling Urge incontinence Gross hematuria Intermittency Severe distention Complete impotence Inadequate erections Stress incontinence
259 215 183 173 161 112 104 103 102 58 49 23 18 16 12 10
(78.5) (65 2) (55.5) (52.4) (48.8) (33.9) (31.5) (31.2) (30.9) (17.6) (14.9) (7.0) (5.5) (4.9) (3.6) (3.0)
Rochester population (person-years)
Initial BPH surgery
Incidence/
(n)
(py)
100,000
Eur Urol 1991;20(suppl 2):11-17
Benign Prostatic Hyperplasia: A Population-Based Study 1605949
C.G. Chute*, W.P. Stephensonb, H.A. Guessh, M. Liebera JMayo Clinic, Rochester. Minnesota, and '’Merck Sharp & Dohme Research Laboratories. West Point. Pennsylvania. USA
Key Words. Prostatectomy • Incidence ■ Epidemiology, population-based Abstract. To evaluate the incidence and outcome of initial surgery for benign prostatic hyperplasia (BPH) and to clarify the natural occurrence and progression of such urologie diseases, two studies have been conducted in a freeliving population in Rochester, MN. The first followed 330 men who had not been diagnosed with prostate or bladder cancer at the time of prostatectomy. All surgery subjects were area residents and between 46 and 95 years of age (mean age 70 years). Among the operated subjects, 14 (4.2%) had serious intraoperative complications, 32 (9.7%) were rehospitalized for urologie complications within 30 days after surgery, and 13 (3.9% ) experienced other serious compli cations in that same time period. Blood transfusions within 30 days of surgery were necessary in 45 patients ( 14% ). The risk of reoperation within óyears of the initial surgery was calculated at 15.1% (95% Cl: 9.7,20.6). On the basis of ageand sex-specific mortality statistics for Rochester, short- and long-term postoperative mortality was not statistically sig nificantly different from that expected. Results of the second study are not yet available. This population-based evalua tion of the natural history of urologie disease is expected to clarify the relative utility of various treatment options and provide a useful perspective on the management of BPH.
tion of a nonreferred study group, and thus provide rel atively unbiased data that derive from a free-living popu lation. Lifetime inpatient and outpatient medical records from the surrounding health care providers were used for Study 1, independent of the patient’s referral to our clinic. The Rochester Epidemiology Project enabled us both to establish a well-defined population base for determining incidence and outcome and to limit our surgical study cohort to men undergoing prostatectomy for the first time and specifically for BPH. This same resource permitted the enumeration of Olmsted County residents around the city of Rochester, from which a random sample could be selected for our natural history cohort (Study 2).
Patients and Methods Study 1: Surgical Intervention for BPH Aprimarytool ofStudy 1 was the Rochester Epidemiology Project, a record-linkage system whereby all medical records from the past sev eral decades have been assembled for Rochester residents [11]. Patient
Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 8:26:52 AM
Benign prostatic hyperplasia (BPH), one of the most common conditions associated with aging in men, has been noted at autopsy in approximately 40% of men in their 50s and in up to 70% of those in their 60s [1]. Reflect ing the high incidence of BPH, an estimated 380,000 trans urethral prostatectomies (TURPs) were performed in the United States in 1987, 97% of which were in men over age 54 [2]. In recognition of the frequency of prostatic sur gery for BPH, a population-based study (Study 1) was conducted to evaluate the incidence and outcome follow ing initial prostatectomy for BPH in a general population, in this case residents of Rochester, MN. As an expansion to this postsurgery study, another population-based cohort (Study 2) from the same geographic area is cur rently being studied to clarify and describe the incidence and natural history of BPH and other urologie diseases in a free-living population. In contrast to most other studies of BPH, which tend to represent highly selected patient groups over a short term [3-10], these two studies are unique in their evalua
Chute/Stephenson/Guess/Lieber
12
Table 1. Characteristics of men having first surgery for benign prostatic hypertrophy in Rochester, MN (1980-1987) Characteristic Age 45-54 55-64 65-74 75-84 85 + Smoking History Current smoker Ex-smoker Never smoked Current nonsmoker (NOS) Unknown Alcohol Use None Occasional/regular Heavy drinker History of alcoholism Unknown Medical History Heart Disease CHD MI Coronary artery bypass/PTCA Congestive heart failure Hypertension Valvular heart disease Congenital heart defect Cerebral vascular disease Transient ischemic attack Stroke Peripheral artery disease IC Gangrene/amputation (vascular insufficiency) Peripheral artery bypass COPD Diabetes mellitus Cancer History of deep vein thrombosis History of pulmonary embolis Bleeding disease
No. Patients (%)
11 89 124 87 19
(3.3) (27.0) (37.6) (26.5) (5.8)
63 164 51 49 3
(19.0) (49.7) (15.5) (14.8) (0.9)
94 190 23 13 10
(28.5) (57.6) (7.0) (3.9) (3.0)
118 (35.8) 103 (31.2) 61 (18.5) 20
(6.0)
28 (8.5) 128 (38.3) 22 (6.7) 2
(0.6)
44 26 31 31 29
(13.3) (7.9) (9.4) (9.4) (8.8)
1 3 80 42 33 10 4
(0.3) (0.9) (24.2) (12.7) (10.0) (3.0) (1.2)
2
(0.6)
NOS = Not otherwise specified; CHD = coronary heart disease; MI = myocardial infarction; PTCA = percutaneous transluminal coro nary angioplasty; IC = intermittent claudication; COPD = chronic obstructive pulmonary disease. From Stephenson et al.12 with permission.
inpatients, on death certificates, and at autopsy. All surgical proce dures performed on Rochester residents at Mayo-affiliated hospitals and at Olmsted Community Hospital also are available in the file. Such a data-retrieval system provides a unique resource for populationbased study. The Study 1 population comprised all men who had undergone prostatectomy for BPH between January 1, 1980. and December 31, 1987. They had been Rochester residents for at least 1 year prior to sur gery, and had not been diagnosed with prostate cancer. Of the 389 men who satisfied the entry criteria, 13 (3.3%) were excluded as a result of histologic (n = 2) or clinical (n=l) evidence of prostate or bladder cancer prior to surgery. Another 46 (12%) were excluded due to prior prostatectomy. A more detailed description of our prostatectomy series (Study 1) methods appears elsewhere [12]. Briefly, collected data included demographics, medical history, urologie history, course and outcome of index surgery and hospitalization, length of hospital stay, subse quent urologie hospitalizations and visits, urologie symptoms within 3(1 days prior to or at any time following surgery (when documented in the medical records), and vital status. Patients’ records were carefully abstracted for evidence of serious intraoperative complications and for surgical mortality and serious surgical morbidity. Annual incidence was calculated as both crude and age-adjusted rates, based on the 1980 figures for the US white male population 45 years of age and over [13]. Changes in incidence with time were analyzed by Poisson regression [14]. Multiple regression analysis was used to evaluate the effects of age and year of surgery on length of hospital stay [15], and logarithmic transformation was used to adjust for the extreme skew of the distribu tions. Risks of reoperation, prostate cancer, and mortality following surgery were analyzed using life table methods (Kaplan-Meier prod uct-limit method) [ 16]. and exact confidence limits for the standardized mortality ratios (SMR) were calculated by Austin’s method [17]. Study 2. Natural History of BPH Study 2 used a population-based survey that randomly sampled 2,000 male residents of Olmsted County, MN, thus drawing from the same population as Study 1. All participants were between the ages of 40 and 80 years. Home visits were conducted and included measure ment of urine flow rates (Dantec 1000 Uroflow, Dantec Elektronik, Skovlunde. Denmark), completion of a validated quality of life and urologie symptom questionnaire, and physician assessment of family history, medication use, and living status [18]. Subsequently, a 500member random subset of these men underwent in-clinic evaluation of prostate size and status via digital rectal examination, standardized prostatic ultrasound, abdominal ultrasound measure of urinary reten tion. prostatic specific antigen, and serum creatinine. Repeat question naire and physical examinations were also conducted at the in-clinic visit. These metrics are expected to be repeated on the 2,000 member cohort 18 to 24 months following enrollment to establish longitudinal change.
records are easily retrievable via a computer-based master file and incorporate outpatient visits to virtually every medical care provider serving the local community (regardless of affiliation with the Mayo Clinic) and cover office consultations, emergency room visits, house calls, and nursing home visits, as well as diagnoses recorded for hospital
Demographic, behavioral, and clinical characteristics of the 330 men in Study 1 are listed in table 1. All of the subjects were at least 45 years of age and 97% were 55 years or older; all but one were white. The most common
Downloaded by: University of Exeter 144.173.6.94 - 6/6/2020 8:26:52 AM
Results
Benign Prostatic Hyperplasia
13
Table 2. Urologie complaints within 30 days prior to surgery for benign prostatic hypertrophy (Study 1)
Table 3. Age-specific incidence of surgery for BPH in Rochester, MN (1980-1987)
Symptom
No. patients (%)
Age
Nocturia Reduced force of stream Frequency Incomplete emptying Hesitancy Complete retention Urgency Dysuria Postvoid dribbling Urge incontinence Gross hematuria Intermittency Severe distention Complete impotence Inadequate erections Stress incontinence
259 215 183 173 161 112 104 103 102 58 49 23 18 16 12 10
(78.5) (65 2) (55.5) (52.4) (48.8) (33.9) (31.5) (31.2) (30.9) (17.6) (14.9) (7.0) (5.5) (4.9) (3.6) (3.0)
Rochester population (person-years)
Initial BPH surgery
Incidence/
(n)
(py)
100,000
