The prostate sits directly beneath the bladder with the urethra running through it. From about the age of forty it tends to grow, and because of where it sits, growth squeezes the tube it surrounds. The result is lower urinary tract symptoms: a weak or hesitant stream, needing to go often and urgently, getting up at night, and the sense of not having emptied.
The first thing to say is what this is not. Benign prostatic hyperplasia is not prostate cancer and does not turn into it. They arise in different parts of the gland. Hyperplasia is a proliferation of smooth muscle and epithelial cells within the transition zone [1], beginning in the periurethral zone [2] — which is why it obstructs. Malignancy develops mainly in the peripheral zone [3], which is why it does not obstruct until late. Both are common in older men and they frequently coexist, which is a coincidence of age rather than a relationship. A man can have either, both, or neither.
It is very common. Prevalence rises after 40 and has been reported anywhere from 8 to 60 percent by age 90 depending on definition [4] [5] [6] [7], and symptoms are commoner still: in a 30,000-person population survey across the USA, UK and Sweden, 72.3 percent of men reported at least one lower urinary tract symptom at least "sometimes", and 47.9 percent at least "often" [8]. The burden falls on quality of life [9] [10], and it travels with sexual dysfunction closely enough that the two are routinely studied together [11] [12] [13] [14] [15].
The second thing to say is that symptoms and gland size are only loosely related. A large prostate can be silent and a modest one can be miserable, because obstruction depends on where the tissue sits and on how the bladder has responded to years of pushing against it. That is why the assessment is built on symptoms rather than on size, and why "how big is it?" is the wrong first question.
Start here: androgens, growth, and what actually obstructs
Prostatic growth depends on dihydrotestosterone, made from testosterone by 5-alpha-reductase, whose isozymes have distinct tissue distributions [16]. That dependence is what makes the gland druggable: block the enzyme and the tissue involutes.
But obstruction has two components, and the distinction organises the entire treatment section. There is a static component — the physical bulk of the tissue — and a dynamic component, the smooth-muscle tone of the prostate and bladder neck, which is under alpha-1 adrenergic control. One is anatomy and the other is tone, and each has its own drug class.
Downstream, the bladder itself changes. Years of voiding against resistance produce detrusor changes that persist, and the natural history of lower urinary tract dysfunction has been followed urodynamically over a decade or more [17] [18]. This matters clinically: relieving the obstruction does not always relieve the symptoms, because some of them are now coming from the bladder rather than the prostate.
Clinical BPH also travels with metabolic syndrome and with inflammation, both of which have been proposed as contributors rather than coincidences [4] [6].
Pillar 1: measurement and diagnosis
Symptoms are the measurement
The International Prostate Symptom Score (IPSS) — seven questions on incomplete emptying, frequency, intermittency, urgency, weak stream, straining and nocturia, each scored 0–5, plus a separate quality-of-life question — is how the disease is quantified [19] [20]. It is the primary endpoint of essentially every trial in this review, and it is also the practical instrument: what decides treatment is not the score alone but how much the patient is bothered by it.
Terminology has been standardised for both nocturia [21] and pressure-flow studies [22].
The objective tests, and what they add
Digital rectal examination estimates size and, more importantly, checks consistency — a hard or nodular gland raises the question of cancer.
Uroflowmetry measures peak flow rate and post-void residual measures what is left behind. Both are informative and neither is decisive: the ICS-'BPH' study examined how uroflowmetry, symptoms and demonstrated bladder outlet obstruction relate to each other [23], and the answer is that they are correlated but far from interchangeable.
PSA, which does two jobs at once
This is where BPH and prostate cancer meet, and it is worth being precise. PSA is made by prostate tissue, benign or malignant — so a large benign gland raises it. That is the central difficulty in interpreting PSA as a cancer test [24] [25], and it contributes to the overdiagnosis problem that dominates the prostate-cancer literature [26].
But in BPH, PSA is doing something useful in its own right: it is a proxy for gland volume, and therefore a predictor of what the disease will do. In the placebo arm of a four-year trial, serum PSA was a strong predictor of future prostate growth [27], and PSA has been shown to be a powerful predictor of acute urinary retention [28]. The same molecule that muddies cancer screening is, in the benign disease, a cheap forecast of progression risk — which is exactly the information needed to choose between the two drug classes below.
When to worry
Most BPH is a quality-of-life problem. The features that change that are acute urinary retention, recurrent urinary infection, bladder stones, visible haematuria, and impairment of kidney function from chronically high bladder pressures. Retention is common enough to have been surveyed worldwide in 6,074 men [29], and the guidelines set out the evaluation [30] [2] [31] [32] [33].
Centerpiece: two drugs, two jobs
The static and dynamic components of obstruction each have a drug class, and MTOPS tested both against each other and together.
The trial randomised 3,047 men to placebo, doxazosin (an alpha-blocker, acting on smooth-muscle tone), finasteride (a 5-alpha-reductase inhibitor, shrinking the gland), or both, with a mean follow-up of 4.5 years. The primary outcome was overall clinical progression: a composite of a ≥4-point rise in symptom score, acute urinary retention, incontinence, renal insufficiency, or recurrent urinary infection [34].
Risk of progression fell 39 percent with doxazosin (P < 0.001), 34 percent with finasteride (P = 0.002), and 66 percent with combination therapy (P < 0.001), the combination being significantly better than either alone [34].
The check the arithmetic was never given. The trial reports those three numbers and never multiplies the first two together. If the drugs acted independently on the same outcome, the combined risk ratio would be the product of the separate ones: (1 − 0.39) × (1 − 0.34) = 0.4026, a 59.7 percent reduction. The observed combination effect was 66 percent — exceeding independent action by 6.3 percentage points. The two drugs are slightly more than multiplicative, which is what you expect of agents working on genuinely different mechanisms rather than competing for the same one.
And the finding that matters more, which the trial states but does not quantify. On the headline composite, doxazosin (39 percent) did better than finasteride (34 percent). But look inside: "the risks of acute urinary retention and the need for invasive therapy were significantly reduced by combination therapy and finasteride but not by doxazosin" [34]. The alpha-blocker beat the 5-ARI on the composite while doing nothing at all for the two hard events inside it.
That is a complete dissociation, and the composite conceals it. The composite is dominated numerically by its symptom-score component — a ≥4-point rise is much commoner than retention or surgery — and the symptom score is precisely what an alpha-blocker moves. Same trial, same patients, opposite ordering depending on which endpoint you read.
The teaching point follows directly. Alpha-blockers relax smooth muscle and work within days, but they do not change the gland, so they do not change where the disease is going. 5-ARIs shrink the gland over months and reduce retention and surgery, but they are slow and their benefit is concentrated in men who have a gland large enough to shrink — prostate volume predicts who responds to finasteride [35], and finasteride's reduction of retention and surgery was demonstrated separately in a four-year trial [36]. So the choice is not which drug is better. It is fast symptom relief, slow risk reduction, or both — and PSA and prostate volume are what tell you whether the second is worth having.
Three honest limits. MTOPS used doxazosin and finasteride; the newer agents in each class (tamsulosin, dutasteride) are not the drugs tested here, though CombAT examined dutasteride with tamsulosin on the same question [37] [38]. The multiplicative comparison assumes the two monotherapy effects are independent estimates of action on one outcome, which is a modelling assumption rather than a trial result. And a 66 percent relative reduction in a composite dominated by a symptom-score threshold is not the same as a 66 percent reduction in anything a patient would call a catastrophe.
Pillar 2: treatment
Treatment is stepwise, and the step is chosen by how much the symptoms bother the man and how likely his disease is to progress [30] [31].
Watchful waiting
For mild or non-bothersome symptoms, doing nothing is a legitimate and common choice, with fluid timing, caffeine and alcohol moderation, and bladder training. Lifestyle measures including exercise and diet have been proposed as genuine strategies rather than filler [4].
Alpha-blockers
Tamsulosin, alfuzosin, doxazosin, terazosin and silodosin relax prostatic and bladder-neck smooth muscle. They work within days to weeks. Tamsulosin was the first selective alpha-1A antagonist studied, in a placebo-controlled trial of 756 patients that showed significant improvement in symptom score and peak flow [39]; silodosin has been trialled similarly [40].
The side effects follow from the mechanism, and they trade off against each other with selectivity. The older, less selective agents block vascular alpha receptors too: terazosin produced significantly more dizziness than tamsulosin in a head-to-head comparison [41], and finasteride carries a lower risk of asthenia, dizziness and postural hypotension than either doxazosin or terazosin [42]. Tamsulosin, being selective, produced orthostatic test results comparable to placebo [39].
What selectivity buys in blood pressure it spends on ejaculation. In a head-to-head trial, abnormal ejaculation was the commonest adverse event with the highly selective silodosin and occurred in 22.3 percent against 1.6 percent with tamsulosin — though only 2.9 percent of men discontinued for it [43], and silodosin's symptomatic benefit has been analysed separately in men with and without retrograde ejaculation [44]. Alpha-blockers are also used to assist a trial without catheter after acute retention [45].
5-alpha-reductase inhibitors
Finasteride and dutasteride block conversion of testosterone to dihydrotestosterone; dutasteride inhibits both isozymes and produces marked DHT suppression [46] [16]. The gland shrinks over six to twelve months, and the benefit is on progression: reduced acute urinary retention and reduced need for surgery [36] [34].
Two practical consequences. First, they lower PSA substantially: in one comparative trial finasteride reduced prostate volume by 18 percent and serum PSA by 41 percent [47]. A man on one of these drugs needs his PSA interpreted against that shift, because a value that looks reassuring may not be. Their own sexual side effects — ejaculation disorder, impotence and lowered libido against placebo — are established [42]. Second, dutasteride was tested for prostate cancer prevention with a result that complicated its use [48], which is a reminder that a drug altering androgen signalling in the prostate does not act only on the benign disease.
Tadalafil
A phosphodiesterase-5 inhibitor taken daily improves both urinary symptoms and erectile function, which is convenient given how often the two coexist [11]. A meta-analysis of eight randomised trials found a mean IPSS improvement of 2.19 points against placebo (P < 0.00001) and an IIEF improvement of 4.66 — but no significant effect on peak urinary flow when doses were pooled, with only the 5 mg dose reaching significance at +0.63 mL/s [49]. It has been compared head-to-head with tamsulosin [50].
That dissociation is worth noticing: tadalafil improves how men feel without measurably improving how they flow, which is another instance of the review's recurring theme — the objective and subjective measures in this disease are not the same measure.
Procedures
Transurethral resection of the prostate (TURP) is the long-standing standard, and its outcomes and complications are exceptionally well characterised — incidence, management and prevention have been reviewed [51], and functional outcomes and complications meta-analysed [52] [53]. Bipolar resection modified the safety profile [54]. Holmium laser enucleation has been compared against TURP in randomised trials with long-term follow-up [55] [56] [57].
The minimally invasive options exist largely because of one side effect. TURP works, but it may lead to sexual dysfunction and incontinence and has a long recovery [58], and for many men retrograde ejaculation is a serious cost. The prostatic urethral lift was tested against TURP in a randomised multinational trial of 80 men using a composite responder endpoint — BPH6 — explicitly built to score symptom relief, quality of recovery, erectile function, ejaculatory function, continence and safety together [58]. Designing the endpoint around what patients mind, rather than around flow rate, is the notable thing about that trial.
Prostatic artery embolisation blocks the blood supply through a catheter, avoiding the urethra entirely. It has been compared against TURP [59] [60], assessed for efficacy and safety [61], followed to medium and long term [62], studied as primary treatment [63], and evaluated on quality of life and urodynamics [64] [65].
Pillar 3: what is unresolved
Choosing among the minimally invasive options. There are now several — urethral lift, water-vapour ablation, aquablation, embolisation — each trading some efficacy for some preservation of function, and the comparative evidence against TURP and against each other is thinner than the enthusiasm [58] [59] [61].
Predicting who progresses. PSA and prostate volume identify men at higher risk of growth and retention [27] [28] [35], which is the basis for offering a 5-ARI. Sharpening that selection would spare the men who will never need it a drug with sexual side effects taken for years.
The bladder, not the prostate. Long urodynamic follow-up shows that lower urinary tract dysfunction in men is not purely obstructive [17] [18], and symptoms that persist after the obstruction is relieved are the clearest evidence. Identifying those men before an operation, rather than after, remains difficult.
Endpoints. The dissociation in MTOPS is not a quirk of one trial — it is what happens when a composite mixes a common soft component with rare hard ones. The IPSS is the field's standard and it is a symptom score; the BPH6 endpoint [58] is one attempt to broaden what gets counted.
Dig deeper in lmmol
Prostate cancer is the essential companion, and the relationship is one of confusion rather than causation: the two diseases share an organ, an age group and a blood test, and nothing else. PSA rises with benign enlargement as well as with malignancy [24] [25], which is a large part of why PSA screening produces overdiagnosis [26] — and in the benign disease that same PSA is a useful predictor of growth and retention [27] [28]. Reading the two reviews together also shows the reverse of the discordance problem: in prostate cancer the difficulty is finding disease that will never cause symptoms, while in BPH it is symptoms that do not track the disease. Kidney stones shares the urinary tract and the acute-obstruction emergency, and both are conditions where the objective finding and the suffering diverge. Chronic kidney disease is the downstream risk when bladder outlet obstruction goes unrelieved long enough to raise pressures upstream. And osteoarthritis is the closest methodological sibling anywhere in this collection: another very common age-related condition where imaging severity and symptom burden come apart, and where treatment is therefore chosen by bother rather than by measurement. The full collection is at health.