Hyperlipidemia patients with long-term statin treatment are associated with a reduced risk of progression of benign prostatic enlargement

Abstract

Objective

To evaluate the impacts of statin treatment on the risk of benign prostatic enlargement (BPE) progression in hyperlipidemia patients.

Methods

Newly diagnosed hyperlipidemia patients (n = 7961), identified from Taiwan’s National Health Insurance Research Database, were divided into four statin cohorts (statin use >365 days, n = 1604; statin use 181–365 days, n = 813; statin use 91–180 days, n = 739; and statin use 31–90 days, n = 713) and one control cohort (cohort that used no statins, n = 4092). Study endpoint was occurrence of BPE progression (BPE diagnosis plus receiving BPE-related medications or surgery). Relative risks of BPE progression in the statin cohorts compared to the control cohort were analyzed.

Results

Multivariable Cox proportional hazards regression analyses demonstrated that BPE progression risk in the cohort used statins for >365 days was significantly lower than the control cohort (adjusted hazard ratio: 0.70, 95% confidence interval: 0.58 ∼ 0.85, p < .001). However, BPE progression risks of the other three statin cohorts did not significantly differ from the control cohort. Trend analysis revealed that the effects of statin treatment on decreasing BPE progression risk were significantly related to statin treatment duration (p = .001).

Conclusions

Hyperlipidemia patients with long-term statin treatment (more than 365 days) are associated with a reduced risk of BPE progression.

Keywords:

• Statin
• prostate
• benign prostate enlargement
• hyperlipidemia
• progression

Introduction

Benign prostatic enlargement (BPE) is common in aging males, and the prevalence of BPE with lower urinary tract symptoms (LUTSs) ranges 50%–75% in males older than 50 years of age [1]. Previous studies indicated that BPE is a progressive condition [2,3]. The BPE progression may result in prostate size enlargement and worsening of LUTSs (especially acute urine retention) [4]. Patients with BPE progression may need to receive BPE-related medications and/or surgery [5]. Preventing the occurrence of BPE progression is important for patients with BPE.

The definitive pathogenesis of BPE progression remains to be determined. Nevertheless, existing evidence supports the concept that the crucial factors in mediating BPE progression may involve genetic, homeostatic, hormonal, inflammatory, and metabolic issues. For instance, patients with BPE progression are noted to have alterations in genetic expression of insulin receptor, insulin-like growth factor, and insulin-like growth factor-binding protein-3 [6]. Loss of prostatic glandular homeostasis between cell renewal and apoptosis, loss of hormonal balance between estrogen and androgen, and alteration in prostatic blood circulation have also been suggested to contribute to the development of BPE progression [7–9].

Of note, inflammation and metabolic factors (e.g. hyperlipidemia) have also been shown to be crucial factors in mediating BPE progression [6–16]. For instance, prostatic inflammation was noted to be associated with prostate enlargement [10]. Dysregulation of the immune response in prostate tissues with T-cell activation and cytokine upregulation was shown to induce prostate growth [17]. In addition, hyperlipidemia can cause prostatic inflammation and was proposed to be an important factor in BPE progression [18]. This concept was confirmed by our recent population-based cohort study, as our data demonstrated that hyperlipidemia patients were associated with a higher risk of BPE progression compared to nonhyperlipidemic patients [19].

As mentioned above, hormone, inflammation, and hyperlipidemia are crucial factors in BPE progression. In line with this notion, it is thus likely that agents with hormone manipulation, lipid-lowering, or anti-inflammatory effects may have the potential to reduce the risk of BPE progression. Statins are potent inhibitors of cholesterol biosynthesis and are widely used in clinical settings as lipid-lowering drugs [20]. Statins also possess testosterone-lowering and potent anti-inflammatory effects [20,21]. Based on these characteristics, we speculated that statins may have the potential to reduce the risk of BPE progression in hyperlipidemia patients. This concept was partially confirmed by previous data. For instance, in a population-based cohort study, statin use was noted to be associated with a reduced risk of BPE and LUTSs [22]. Data from a randomized prospective study with a total of 124 patients demonstrated that statin use could reduce the prostatic volume and improve LUTSs in patients with metabolic syndrome [23]. However, large-scale data in this regard have not been evaluated to date. To further elucidate this issue, we conducted this population-based longitudinal cohort study with the hypothesis that statin use could reduce the risk of BPE progression in hyperlipidemia patients.

Materials

Ethics

This study complies with the Personal Information Protection Act in Taiwan and was approved by the Institutional Review Board of Taipei Tzu Chi Hospital (protocol no.: 06-W02-018). Because the National Health Insurance (NHI) Research Database (NHIRD) consists of anonymous data, no patient consent is acquired.

Database

This study was conducted using data from the NHIRD, an insurance claims database released by the National Health Research Institutes for research purposes in Taiwan. The NHIRD is a database containing comprehensive longitudinal computerized medical records of all people in the NHI program. The NHI program covers approximately 99% of the Taiwanese population (approximate 23.5 million). Data on one million randomly selected individuals from all beneficiaries of the NHIRD were used to create the Longitudinal Health Insurance Database 2010 (LHID2010). There were no significant differences in the age, sex, or average insured payroll-related amount between individuals in the NHIRD and those in the LHID2010. LHID2010 contains patient demographics, diagnostic data, medical services received, and drug codes. The diagnostic accuracy of the NHI claims data with International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic codes has been validated [24]. This study employed the ICD-9-CM codes to identify subject’s medical diagnosis (Table 1). This study also employed the anatomical therapeutic chemical (ATC) classification system codes and the Taiwan NHI reimbursement procedure codes to identify subject’s exposure to medications and procedures, respectively (Table 1).

Table 1. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, the Taiwan National Health Insurance reimbursement procedure codes, and the anatomical therapeutic chemical (ATC) classification system codes used in the study.

Diagnosis/Procedure/Medication	Codes
Diagnosis	ICD-9-CM
Hyperlipidemia	272.x
Benign prostate hyperplasia	600.xx
Diabetes mellitus	250, 648.0
Chronic hypertension	401–405
Coronary heart disease	410.xx-414.xx
Obesity	278, 278.0, 278.00, 278.01
Liver cirrhosis	571.2, 571.5, and 571.6
Procedure	Procedure
Transurethral resection of the prostate	79406B (resection prostate chips of >5 g), 79411B (15–50 g), 79412B (>50 g)
Medication	ATC
Statins	C10AA
Nonsteroidal anti-inflammatory drugs	M01A
Metformin	A10BA02
Aspirin	B01AC06, B01AC30, N02BA01, and N02BA51
Lipid-lowering drugs except statins	C10AB, C10AC, C10AD, C10AX, C10BA, and C10BX
Alpha-blockers	G04CA
5-alpha reductase inhibitors	G04CB01, G04CB02

Study design and study endpoint

Hyperlipidemia patients were identified with at least two separate in- or outpatient medical records between January 2000 and December 2012, indicating a new diagnosis of hyperlipidemia. Hyperlipidemia patients with statin use were identified with the ATC classification system code. Physicians in Taiwan adhere to the diagnostic criteria of hyperlipidemia and prescribing guidelines for statins provided by the NHI Administration, Ministry of Health and Welfare, Taiwan [25].

Hyperlipidemia patients were divided into four statin cohorts (i.e. statin use for >365 days, for 181–365 days, for 91–180 days, and for 31–90 days) and one control cohort (i.e. cohort that used no statins). The first prescription date for statins was used as the index date for hyperlipidemia patients with statin use (i.e. the statin cohorts). The statin prescription duration after the index date was defined as the statin treatment duration. The date of the hyperlipidemia diagnosis was used as the index date for hyperlipidemia patients who did not use statins (i.e. the control cohort). The study endpoint was the occurrence of BPE progression (i.e. a BPE diagnosis plus use of alpha-blockers or 5-alpha reductase inhibitors [5ARIs] or receiving transurethral resection of the prostate [TURP]).

Patients of a female gender, who were aged younger than 40 years or older than 100 years, had a diagnosis of hyperlipidemia on only one occasion, or had a diagnosis of BPE before the diagnosis of hyperlipidemia or a diagnosis of prostate cancer, were excluded. Patients who received statins before the diagnosis of hyperlipidemia received statins during admission or received statins for less than 31 days were also excluded. The confounders considered in this study were identified within 1 year before the index date. Comorbidities related to BPE progression considered in this study were diabetes mellitus (DM) [26], hypertension [27], coronary heart disease (CHD) [28], obesity [29], and liver cirrhosis [30]. Medications related to BPE progression were nonsteroidal anti-inflammatory drugs (NSAIDs) [31], metformin [32], aspirin [31], and lipid-lowering drugs except statins [33]. Potential surveillance bias can occur due to the frequency of health checkups among the five cohorts, and so urologist visits were recorded and included as a confounding factor.

Statistical analysis

The SPSS statistical package (version 21.0, SPSS, IBM Corporation, Somers, NY) was used to analyze all data. Continuous data were presented as mean ± standard deviation. Differences among the five cohorts were measured by the chi-squared test or the Kruskal–Wallis H-test. A p value of < .05 was considered statistically significant. The number needed to treat (NNT) was calculated according to previously published model [34]. The relative risks of having BPE progression in the statin cohorts compared to the control cohort were estimated by Cox proportional hazard regressions. The trend analysis regarding the effects of duration of statin use on the risk of having BPE progression was also performed using Cox proportional hazards regression. Multivariable Cox models were used to adjust for possible confounders that might influence the risk of BPE progression. In addition, a multiple logistic regression was conducted to estimate a propensity score, in which statin use was the dependent variable and confounders were independent variables. The propensity score was further adjusted in the Cox regression analyses.

Results

Characteristics of study subjects

In total, 7961 patients with newly diagnosed hyperlipidemia were identified. As mentioned above, hyperlipidemia patients were divided into four statin cohorts (i.e. with statins use for >365 days, n = 1604; for 181–365 days, n = 813; for 91–180 days, n = 739; and for 31–90 days, n = 713) and the control cohort (n = 4092) (Figure 1). The five cohorts significantly differed in age (p < .001; Table 2). The five cohorts also significantly differed in the incidences of DM, hypertension, CHD, metformin use, aspirin use, other lipid-lowering drug use, and urologist visits (all p < .05; Table 2). In contrast, the incidences of obesity, liver cirrhosis, and NSAID use did not significantly differ among the five cohorts (Table 2).

Figure 1. Flowchart of the study design. LHID: Longitudinal Health Insurance Database; BPE: benign prostate enlargement.

Table 2. Distribution of sample characteristics according to the duration of statin use in patients with hyperlipidemia.

	Statin use
	>365 days	181–365 days	91–180 days	31–90 days	None
	(n = 1604)	(n = 813)	(n = 739)	(n = 713)	(n = 4092)	p^a
Age, n (%)						<.001
40–49	693 (43.2)	375 (46.1)	348 (47.1)	341 (47.8)	2038 (49.8)
50–59	560 (34.9)	273 (33.6)	255 (34.5)	250 (35.1)	1315 (32.1)
60–69	266 (16.6)	115 (14.1)	99 (13.4)	81 (11.4)	497 (12.1)
70–99	85 (5.3)	50 (6.2)	37 (5)	41 (5.8)	242 (5.9)
Diabetes mellitus, n (%)	291 (18.1)	104 (12.8)	98 (13.3)	62 (8.7)	406 (9.9)	<.001
Hypertension, n (%)	556 (34.7)	242 (29.8)	208 (28.1)	159 (22.3)	901 (22)	<.001
Coronary heart disease, n (%)	144 (9)	62 (7.6)	52 (7)	52 (7.3)	239 (5.8)	.001
Obesity, n (%)	5 (0.3)	6 (0.7)	5 (0.7)	2 (0.3)	33 (0.8)	.193
Liver cirrhosis, n (%)	4 (0.2)	2 (0.2)	5 (0.7)	3 (0.4)	20 (0.5)	.514
NSAID use, n (%)	1173 (73.1)	611 (75.2)	547 (74)	523 (73.4)	2902 (70.9)	.055
Metformin use, n (%)	108 (6.7)	27 (3.3)	35 (4.7)	20 (2.8)	91 (2.2)	<.001
Aspirin use, n (%)	273 (17)	112 (13.8)	109 (14.7)	94 (13.2)	470 (11.5)	<.001
Lipid-lowering drugs except stains, n (%)	299 (18.6)	128 (15.7)	126 (17.1)	133 (18.7)	899 (22)	<.001
Urologist visits (mean ± SD)	3.06 ± 8.39	2.62 ± 9.09	2.34 ± 6.27	2.02 ± 6.16	2.86 ± 8.41	<.001

^a Evaluated by the chi-squared test or the Kruskal–Wallis H-test.

NSAID: nonsteroidal anti-inflammatory drug; SD: standard deviation.

Risk of BPE progression in hyperlipidemia patients

The incidences of BPE progression in the cohort with statin use for >365 days (8.3%), the cohort with statin use for 181–365 days (8.6%), the cohort with statin use for 91–180 days (7.7%), the cohort with statin use for 31–90 days (7.2%), and the control cohort (11.3%) differed significantly (p < .001). In addition, the mean time for the occurrence of BPE progression in the cohort with statin use for >365 days (4.67 ± 3.00 years), the cohort with statin use for 181–365 days (3.30 ± 2.38 years), the cohort with statin use for 91–180 days (3.04 ± 2.56 years), the cohort with statin use for 31–90 days (2.32 ± 2.44 years), and the control cohort (3.83 ± 3.00 years) also significantly differed (p < .001).

Our data further revealed that the risk of having BPE progression in the cohort with statin use for >365 days was significantly lower than the control cohort (hazard ratio [HR] = 0.70, 95% confidence interval [CI]: 0.58–0.85, p < .001; Table 3). The risk of having BPE progression in the cohort with statin use for >365 days remained significantly lower than the control cohort after adjusting for potential confounders (adjusted HR =0.66, 95%CI: 0.54–0.8, p < .001; Table 3) and propensity score (adjusted HR =0.68, 95%CI: 0.56–0.82, p < .001; Table 3). Comparing to the control cohort, the NNT for preventing one BPE progression in the cohort with statin use for >365 days was 32.82. To test the robustness of our results, we further stratified our analyses according to the baseline confounders. The results indicated that, compared with the control cohort, the cohort with statin use >365 days was associated with a lower risk of BPE progression in all subgroups stratified by confounding factors (Figure 2). As the case number of obesity and liver cirrhosis was small, we thus did not include these two factors into stratified analysis.

Figure 2. The risk of having benign prostate hyperplasia progression in the cohort with statin use >365 days compared to the cohort that used no statins stratified by confounders. DM: diabetes mellitus; CHD: coronary heart disease; NSAIDs: nonsteroidal anti-inflammatory drugs; HR: hazard ratio; CI: confidence interval.

Table 3. Statin use duration in predicting benign prostate enlargement progression in hyperlipidemia patients.

Variable	HR	95%CI	p-Values*
Statin use (>365 days)	0.70	0.58–0.85	<.001
Statin use (181–365 days)	0.98	0.77–1.27	.899
Statin use (91–180 days)	0.99	0.75–1.31	.944
Statin use (31–90 days)	1.07	0.80–1.42	.672
Statins use (none)	1.00
Trend			.001
Statin use (>365 days)^a	0.66	0.54–0.81	<.001
Statin use (181–365 days)^a	0.92	0.72–1.19	.542
Statin use (91–180 days)^a	1.00	0.76–1.33	.978
Statin use (31–90 days)^a	1.13	0.85–1.52	.405
Statin use (none)	1.00
Trend^a			<.001
Statin use (>365 days)^b	0.68	0.56–0.82	<.001
Statin use (181–365 days)^b	0.96	0.75–1.24	.774
Statin use (91–180 days)^b	0.97	0.74–1.28	.836
Statin use (31–90 days)^b	1.05	0.79–1.41	.725
Statin use (none)	1.00
Trend^b			<.001

HR: hazard ratio; CI: confidence interval.

* Evaluated by the Cox proportional hazard regression.

^a Adjusted for age, diabetes mellitus, hypertension, coronary heart disease, metformin use, aspirin use, use of lipid-lowering drugs except statins, and urology visit.

^b Adjusted for propensity score.

Of note, the risks of having BPE progression in the cohorts with statin use for 181–365 days, with statin use for 91–180 days, and with statin use for 31–90 days did not significantly differ from the control cohort (Table 3). In addition, the trend analysis revealed that the effects of statin use on decreasing the risk of BPE progression in the study cohorts were significantly related to the duration of statin use (p = .001; Table 3). The trends remained the same after adjusting for potential confounders (p < .001; Table 3) and propensity score (p < .001; Table 3).

Discussion

Inflammatory, metabolic, and hormonal factors are important factors for BPE progression [6,8,12,13,16,17]. Hyperlipidemia induces prostatic inflammation, so it is reasonable to observe an association between hyperlipidemia and BPE progression [19,33]. As statins possess potent lipid-lowering and anti-inflammation capacities, we thus hypothesized that statin use might decrease the risk of BPE progression in hyperlipidemia patients. Findings from this study confirmed our hypothesis and demonstrated for the first time that hyperlipidemia patients treated with statins for more than 365 days had a 30%–34% decrease in the risk of experiencing BPE progression (i.e. receiving BPE-related medications or TURP) compared to those hyperlipidemia patients who used no statins.

As mentioned above, several factors can influence BPE progression. For instance, previous reports indicated that patients of an advanced age [28], and with DM [26], hypertension [27], CHD [28], and/or obesity [35] are at an increased risk of BPE progression. Liver cirrhosis patients, on the contrary, were reported to have a lower rate of BPE compared to the general population [30]. In addition, NSAIDs [31], aspirin [31], metformin [32], and lipid-lowering drugs [33] were reported to decrease the risk of prostate enlargement. Data from this study revealed that some of these confounders significantly differed in these five cohorts. Possible impacts of these confounding factors were thus controlled for using a multivariable Cox proportional hazard regression model and propensity score analysis in this study. After controlling for these potential confounders, our data revealed the same trend and demonstrated a reduced risk of having BPE progression in hyperlipidemia patients treated with statins for more than 365 days. The NNT in association with one patient free of BPE progression was 32.82 for patients used statins more than 365 days. Our data also demonstrated that the protective effect of statins was related to the treatment duration. Collectively, these data highlight the beneficial potential of long-term statin use (more than 365 days) on decreasing the risk of BPE progression in hyperlipidemia patients.

Data from this study revealed that more than half of the included hyperlipidemia subjects (52.2%) did not receive statin treatment. In addition, among statin-treated hyperlipidemia patients, only 39.9% of them received long-term statin treatment (>365 days). The reason why physicians did not prescribe statins or did prescribe statins but for only a short period of time (e.g. less than 365 days) for the included hyperlipidemia patients remains to be elucidated. Of note, findings from this study demonstrate that the protective effect of statins is related to the treatment duration. Data from previous reports seem to support this notion [36,37]. Previous data demonstrated that the therapeutic effects of statins on reducing the risk of myocardial infarction and stroke in hyperlipidemia patients were only seen in those who received statin treatment for a treatment duration of at least 1 year [36,37]. As both the formation of atherosclerotic lesions and BPE progression in hyperlipidemia patients were associated with chronic inflammation [33], it is thus reasonable to observe that the beneficial effects of statins were only seen in those hyperlipidemia patients with long-term statin use (e.g. >365 days).

In addition to the treatment duration, treatment adherence also affects the therapeutic effects of treatment. For instance, 5ARIs are effective drugs for decreasing the risk of urine retention and receiving a BPE-related operation in patients with a large prostate [38,39]. Persistent use of 5ARIs is a key factor in preventing BPE progression, as an increased prostatic volume and exacerbation of voiding symptoms will occur if treatment with 5ARIs is discontinued [40,41]. One previous population-based study reported that the incidence of patients with good statin adherence was approximately 40.8% in Taiwan [42]. This study did not investigate statin adherence. However, judging from our data that hyperlipidemia patients who received long-term statin treatment (i.e. >1 year) accounted for approximately 39.9%, these data seem to suggest that these patients may have better statin adherence than those who received statin treatment for less than 1 year. If so, then it is reasonable to observe beneficial effects of long-term statin treatment on reducing the risk of BPE progression in hyperlipidemia patients.

Although this study shows the beneficial effects of long-term statin use on reducing BPE progression in hyperlipidemia patients, there are some important limitations of our study. First, nutrition and dietary factors are important for prostatic disease progression [43–45], information about personal behaviors (e.g. alcohol consumption, smoking, dietary habits, and physical exercise), lifestyle, and family history were lacking in the NHIRD. Second, direct evidence to delineate a cause–effect relationship between statin use and BPE progression is still lacking. Large-scale prospective randomized clinical studies are required before further conclusions can be drawn. Third, different types of statins may have different effects on preventing BPE progression. However, the individual effects of different types of statins were not analyzed in this study. Fourth, detailed information about clinical symptoms, the International Prostate Symptom Score, prostate-specific antigen level, and image findings are important indicators for TURP [46]. These indicators are unavailable in the NHIRD. Fifth, adverse events may occur in some patients undergoing statin therapy. A risk–benefit assessment is suggested if further interpretation of our data is intended.

Conclusions

Data from this population-based longitudinal cohort study show the potential of long-term statin use (more than 365 days) to reduce the risk of BPE progression in hyperlipidemia patients. The protective effect of statins is related to the treatment duration.

Acknowledgements

Part of the study results had been presented at the 33rd European Association of Urology Annual Congress, March 16–20 2018, Copenhagen, Denmark.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Wan Fang Hospital under Grant 107-wf-eva-07 awarded to HJ Shih and CJ Huang.