The efficacy of PDE5 inhibitors in diabetic patients

Abstract

Background

Phosphodiesterase 5 inhibitors (PDE5i), since their introduction in the late 1990s, have proven their efficacy in treating several conditions, predominantly pulmonary hypertension and erectile dysfunction where they remain the first‐line therapeutic option. However, in the recent years, growing evidence from both animal and human studies has emerged to suggest the additional benefits of PDE5i in cardiovascular and metabolic disorders. This is of specific interest to the diabetes population where prevalent cardiovascular disease and metabolic dysregulation significantly contribute to the increased morbidity and mortality.

Objectives

To examine the available data on the non‐standard, pleiotropic effects of PDE5i in patients with diabetes mellitus.

Materials and methods

The review of the published background research, preclinical studies and clinical trials.

Results

In human studies, PDE5 inhibition appeared to be associated with reduced cardiovascular mortality and overall improved clinical outcomes in those with established cardiovascular disease. PDE5i were also consistently found to reduce albuminuria in subjects with diabetic nephropathy. Furthermore, animal data suggest a plausible effect of this group of medication on sensory function and neuropathic symptoms in diabetic neuropathy as well as improved wound healing. A decrease in insulin resistance and augmentation of beta cell function seen in preclinical studies has not been consistently demonstrated in human trials.

Discussion and conclusion

In animal models, PDE5 inhibition appears to decrease oxidative stress and reduce some of the micro‐ and macrovascular complications associated with diabetes. However, data from human trials are limited and largely inconsistent, highlighting the need for adequately powered, randomised‐controlled trials in diabetic cohorts in order to fully assess the benefits of PDE5i in this group of patients.

Phosphodiesterases (PDEs) are a family of enzymes which hydrolyse cyclic nucleotides, such as cyclic adenosine 3′,5′‐monophosphate cAMP) and/or cyclic guanosine 3′,5′‐ monophosphate (cGMP), into their inactive 5′‐monophosphate form, thus regulating the duration and intensity of various intracellular functions. ¹ Cyclic GMP is synthesised from guanosine‐5′‐triphosphate (GTP) in response to nitric oxide (NO) generated by the conversion of l‐arginine by NO synthase. Cyclic GMP activates cGMP‐dependent protein kinase G (PKG) which evokes a reduction of intracellular Ca²⁺ concentration resulting in vasodilation. The downstream effects of NO are, however, limited by the degradation of cGMP by phosphodiesterase type 5 (PDE5). The enzyme, localised subcellularly within the centrosomal and vesicle compartments of the vascular smooth muscle, is highly expressed in human corpora cavernosa, cerebellum, kidney and lung. ² , ³ Given the pivotal role of PDE5 in the circulatory regulation and smooth muscle tone, the research has focused on investigating the consequences of the PDE5 inhibition which leads to the accumulation of cGMP and potentiates the vasodilatory effect of NO, particularly in the conditions associated with endothelial dysfunction and impaired microcirculation. Several studies have reported that NO‐mediated vasodilatation is abnormal in type 2 diabetes mellitus (T2DM), ⁴ presumed to be indeed a consequence of endothelial dysfunction and chronic inflammation, both leading to micro‐ and macrovascular complications.

1. PDE5 INHIBITORS AND CARDIOVASCULAR DISEASE IN DIABETES

Endothelial dysfunction and a consequent reduction of NO‐dependent vasodilation have been implicated in the aetiology of both T2DM and cardiovascular disease. Nonetheless, diabetes with its metabolic consequences is itself considered a cardiovascular risk factor increasing, among others, myocardial susceptibility to ischemia/reperfusion injury (characterised by deficient oxygen supply and subsequent restoration of blood flow) and, by disrupting intracellular signalling pathways, reducing resistance of cardiomyocytes to cell death. ⁵ In fact, patients with diabetes tend to develop heart failure after ischemia/reperfusion injury at more than double the rate of non‐diabetics, and they have reduced life expectancy than the average. ⁵

It is well documented that cGMP and PKG attenuate stress responses in the heart reduce pathological hypertrophy and enhance cardiac cell contractility, survival and mitochondrial function. ⁶ Because the PDE5 enzyme is constitutively expressed in cardiac muscle cells, in addition to vasculature, it has been proposed that the PDE5 inhibitors (PDE5i) might prove useful in restoring the NO/cGMP/PKG pathway in diseased cardiomyocytes, improving cardiac function and overall survival of patients with T2DM.

Certainly in non‐diabetic animal models, sildenafil and vardenafil were cardioprotective in ischaemic heart disease, reduced susceptibility to cardiac arrhythmias and improved cardiac function. ⁷ , ⁸ , ⁹ In diabetic mice, tadalafil treatment modulated and upregulated proteins involved in cytoskeletal rearrangement and cardiac remodelling enhanced antioxidant enzymes and attenuated mitochondrial dysfunction. ¹⁰ , ¹¹ PDE5i also reduced the size of infarct post‐ischemia/reperfusion injury as evidenced by lower rates of cardiomyocyte apoptosis and necrosis, accompanied by reduction in proinflammatory cytokine levels. ¹² Moreover, PDE5 inhibition with long‐term vardenafil introduced in the early stages of diabetes prevented the development of T2DM ‐associated left ventricular dysfunction and restored the activity of cGMP–PKG axis by increasing myocardial as well as plasma cGMP levels. ¹³

In addition to the above pleiotropic effects, the inhibition of PDE5 has been linked to cardiac adrenergic signalling regulation. ¹⁴ In cardiomyocytes, PDE5 exists in a protein complex with β2 adrenergic receptor (β2AR), and the functional association of the two contributes to reduced cardiac contractility during the development of diabetic cardiomyopathy. In animal models, chronic PDE5 inhibition with sildenafil improved the adrenergic stimulation of cGMP signals and enhanced excitation–contraction (E–C) coupling of diabetic myocytes leading to improved cardiac function both at baseline and after adrenergic stimulation. ¹⁵

The data from the human studies of the effects of PDE5i on cardiovascular health and parameters largely come from populations with pre‐existing cardiovascular disease not specifically stratified for their diabetes status; the studies performed in diabetic populations only are limited.

Acute administration of PDE5i to patients with known atherosclerosis, ¹⁶ coronary artery stenosis, ¹⁷ coronary heart disease, ¹⁸ congestive heart failure ¹⁹ or chronic stable angina ²⁰ not only did not demonstrate any adverse cardiovascular effects but also resulted in improved endothelial function in the coronary circulation, ¹⁶ increased total exercise time in heart failure patients ¹⁹ and did not counteract beneficial effects of concurrently used beta‐blockers. ²⁰ In patients with heart failure with reduced ejection fraction, chronic PDE5i use was associated with improved functional status, including exercise tolerance. ²¹ , ²² In another cohort of patients with heart failure but with preserved ejection fraction cardiac haemodynamics, left ventricular diastolic function, right ventricular systolic function and left ventricular hypertrophy all improved with 6–12 months of therapy with a PDE5i, as compared with placebo. ²³

In a randomised clinical trial designed to investigate the putative anti‐remodelling effect of PDE5i, a 12‐week administration of sildenafil at 100 mg/day or placebo to 59 men with non‐ischaemic diabetic cardiomyopathy improved cardiac kinetics and performance, including cardiac output and ejection fraction, independent of endothelial or metabolic factors. ²⁴ These findings, however, were not confirmed by a larger multicentre‐, double‐blind, placebo‐controlled, parallel‐group, randomised RELAX (the PDE‐5 inhibition to improve clinical status and exercise capacity in heart failure with preserved ejection fraction) trial of 216 patients with heart failure whose exercise capacity and clinical status did not change the following 6‐month period of daily sildenafil administration. ²⁵ However, it might not be possible to generalise these findings to the diabetes population as the study sample was heterogeneous with just over 40% of participants suffering from diabetes.

Further beneficial effects of PDE5i have been reported in cohorts of patients with underlying coronary artery disease (CAD). In clinical trials in male subjects with stable CAD, an administration of a single dose of sildenafil or vardenafil just before the exercise tolerance test significantly prolonged time to ischemic threshold, ²⁶ total exercise duration and times to onset of angina and limiting angina. ²⁷ In female subjects with known ischaemic heart disease/microvascular coronary dysfunction, acute administration of 100 mg sildenafil resulted in improved coronary flow reserve. ²⁸ However, coronary flow reserve did not significantly improve following an acute administration of sildenafil in 21 diabetic men with known erectile dysfunction and asymptomatic CAD. ²⁹

Finally, PDE5 inhibition appears to play a role in reducing cardiovascular mortality in T2DM. In a large observational study of 5956 men with T2DM followed for a median period of 7 years, PDE5i use was associated with a significantly lower incidence of acute myocardial infarction and lower mortality in those who experienced acute myocardial infarction during the study period compared to those who did not use PDE5i. In addition, PDE5i users have significantly lower all‐cause mortality which persisted after adjustment for known risk modifiers, including, among others, age, hypertension and β‐blocker and statin use. ³⁰ Similarly, in a smaller study of 291 type 2 diabetic men with silent, angiographically documented CAD, PDE5i and statin use was associated with a lower incidence of major adverse cardiac events during the median follow‐up of 4 years. ³¹

Taken together, there is some evidence from preclinical and clinical studies demonstrating improved outcomes and the cardioprotective role of PDE5 inhibition in patients with established cardiovascular disease. However, the generalisability of the studies, done predominantly in unselected as opposed to diabetic‐only cohorts, to the diabetes populations must be carefully considered.

2. ANTIOXIDANT AND ANTI‐APOPTOTIC CELLULAR EFFECTS OF PDE5I

Recent in vitro and in vivo studies indicate that PDE5i may have a plausible effect on cell survival, possibly through an NO‐dependent pathway, as well as by directly influencing the synthesis of the key factors regulating the process of apoptosis.

In addition to the earlier described PDE5i effect on limiting necrosis and apoptosis of adult cardiac myocytes in the context of cardiac ischaemia, ¹² in experimental animal models sildenafil also attenuated drug‐induced nephrotoxicity not only through an increase in NO but also through an increase in antiapoptotic B‐cell lymphoma 2 (Bcl‐2) protein expression and decrease in proapoptotic Bax protein expression. ³² In mice post‐cavernous nerve injury, tadalafil decreased apoptosis in the cavernous endothelial and smooth muscle cells and increased the activation of cell survival‐associated kinases, a finding which supports early pharmacological intervention with PDE5i post‐radical prostatectomy. ³³

In the diabetic animal models, the antiapoptotic effects of PDE5i have been studied predominantly in the corpora cavernosa where the decreases in smooth muscle and endothelium density as well as increased number of apoptotic cells were observed. ³⁴ Similar was noted in a small group of diabetes patients who, compared to non‐diabetic controls, had increased apoptotic cell density at vascular sites with impaired cell proliferation which correlated further with poor response to intracavernosal vasoactive substance injection. ³⁵

Evidence suggests that prolonged treatment of aged diabetic rats with sildenafil or tadalafil, with or without additional testosterone supplementation, results in an improvement in cavernosal apoptotic indices, namely increased Bcl₂, cGMP, Bcl₂/Bax ratio and decreased Bax. ³⁶

Another mechanism in which PDE5i could be displaying their antiapoptotic effect in diabetes is through the reduction of oxidative stress. The increased production of reactive oxygen species (ROS), free radicals and advanced glycation products in diabetes has been well recognised and linked to the development of diabetic complications. ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴²

In preclinical studies, daily treatment of diabetic rats with sildenafil resulted in improvement in renal function and reduction in levels of markers of oxidative stress such as malondialdehyde and inhibition of pro‐inflammatory cytokines. ⁴³ , ⁴⁴ Sildenafil appeared to have a modulatory effect on the pro‐inflammatory high‐mobility group box 1 (HMGB1)/nuclear factor‐kappa b (NF‐κB) pathway. ⁴⁴ Concurrently, PDE5 inhibition demonstrated glucose lowering effects, likely through upregulating cGMP levels and thus improving insulin sensitivity.

In recent years, further evidence emerged suggesting that the tissue response to PDE5i might also be mediated via the endogenous antioxidant heme oxygenase (HO) system which is emerging as an important component of the diabetes‐related inflammation. ⁴⁵ Although its main role has been linked to heme catabolism, the upregulation of HO1 isoform was previously shown to be protective against ischaemia and hypoxia. ⁴⁶ , ⁴⁷ In preclinical studies in rodents with experimentally induced diabetes, HO1 upregulation decreased ROS, restored NOS and as well as reduced glomerular injury and apoptosis. ⁴⁸ , ⁴⁹ , ⁵⁰ PDE5i have been trialled in this context in a limited number of preclinical studies, mainly in the rodent models of renal ischaemia‐reperfusion injury, and demonstrated renoprotective properties owing to an induction of overexpression of HO1 antioxidant and antiapoptotic genes and attenuation of proinflammatory cytokines. ⁵¹ , ⁵²

3. PDE5 INHIBITORS AND GLUCOSE METABOLISM DISORDERS

Insulin resistance is a significant risk factor associated with cardiovascular disease, obesity, and a feature of type 2 diabetes, where the ability of insulin to stimulate glucose uptake by insulin‐sensitive tissues is impaired.

It has been established that NO plays an important role in mediating the metabolic effects of insulin, including the stimulation of muscle glucose uptake. ⁵³ Animal models lacking endothelial NO synthase are insulin resistant. ⁵⁴

Nitric oxide synthase and PDE5 have been shown to be highly expressed in tissues involved in glucose metabolism regulation, such as skeletal muscle cells, hepatocytes and endothelial cells. Research suggests that PDE5i could target those cells and act as insulin sensitiser. ⁵⁵ , ⁵⁶ , ⁵⁷ In human skeletal muscle cells, a single dose of tadalafil induced, among others, an increase of glucose transporter (GLUT) 4 gene expression, activation of the peroxisome proliferator‐activated receptor gamma (PPARc), stimulation of the phosphorylation of insulin receptor substrate 1, mammalian target of rapamycin (mTOR) and protein kinase B. ⁵⁸ The latter can be activated by NO and cGMP and plays an important role in insulin‐induced effects, including glucose uptake and GLUT4 translocation. ⁵⁹ Moreover, in human endothelial cells in insulin resistance conditions administration of sildenafil activated NOS and increased NO production, thus reducing the oxidative stress triggered by hyperglycaemia. ⁶⁰

PDE5 is also highly expressed in adipocytes, and studies have shown that PDE5i promote adipogenesis, decrease inflammatory markers’ expression and enhance insulin sensitivity in those cells, largely by improving their mitochondrial function. ⁵⁶

These observations have further been investigated in in vivo studies. For example, sildenafil treatment administered to high fat diet‐fed mice resulted in lower fasting glucose and insulin levels as well as enhanced muscle glucose uptake. ⁶¹ In another study, hepatic inflammation and insulin resistance induced by high fat feeding were fully blocked by PDE5 inhibition. ⁶²

In a small randomised, double‐blind, placebo‐controlled clinical trial in 618 non‐diabetic men with obesity, tadalafil administration for 28 days did not modify insulin secretion or insulin sensitivity. ⁶³ However, in another randomised, double‐blind, placebo‐controlled study of 42 overweight individuals with prediabetes, 12 weeks treatment with sildenafil significantly increased insulin sensitivity index whilst there was no difference between acute‐ or late‐phase glucose‐stimulated insulin secretion between the sildenafil‐ and placebo‐treated participants. ⁶⁴ Similarly, acute treatment with tadalafil induced muscle glucose uptake in a fasted and post‐prandial state in a small sample of subjects with T2DM. ⁶⁵ , ⁶⁶ The fasting glucose and insulin levels have, however, remained unchanged. Ho et al. conducted a randomised, double‐blinded, placebo‐controlled trial of 53 adults with obesity and elevated fasting insulin levels. After 3 months of treatment with tadalafil, an improvement in insulin resistance was observed only in individuals with severe obesity (BMI ≥ 36.2 kg/m²) but not others. ⁶⁷

4. PDE5 INHIBITION REDUCES ALBUMINURIA IN SUBJECTS WITH DIABETIC NEPHROPATHY

Diabetic nephropathy is the leading cause of end‐stage renal disease worldwide. It is characterised by the development of proteinuria, which is thought to be a consequence of hyperglycaemia‐induced oxidative stress and podocyte damage, followed by a decline in glomerular filtration in association with glomerulosclerosis. ⁶⁸ The NO‐cGMP axis plays an important role in the maintenance of renal perfusion and glomerular filtration, with a key role in modulating the slit membrane and cytoskeletal organisation of podocytes. ⁶⁹ , ⁷⁰ Reduced bioavailability and/or uncoupling of NO with a consequent reduction in cGMP levels in the kidney has been proposed as a mechanism contributing to the progression of diabetic nephropathy. ⁷¹ , ⁷² These renoprotective effects and improvements in kidney function are possibly a result of both haemodynamic and intrarenal anti‐inflammatory, anti‐proliferative and antioxidant mechanisms of PDE5i action. ⁷³

Interestingly, in early diabetic nephropathy, an enhanced intrarenal NO production has been observed, which may contribute to hyperfiltration and microalbuminuria. However, the advanced nephropathy, typically associated with proteinuria, declining renal function and hypertension, is recognised as a state of progressive NO deficiency. Hyperglycaemia, accumulation of advanced glycosylation end products and increased oxidative stress appear to be some of the contributory factors leading to decreased NO production and/or availability. ⁷⁴ , ⁷⁵

There is evidence from preclinical studies that the restoration of the NO pathway through PDE5 inhibition may ameliorate oxidative and inflammatory renal damage with consequent reductions in albuminuria, glomerular hyperfiltration and hypertrophy and overall reduction in glomerulosclerosis. ⁷⁶ , ⁷⁷ The inhibition of PDE5, which is expressed in the proximal tubules, collecting ducts and glomerulus, appears to directly attenuate podocyte damage by raising an intracellular cGMP content of podocytes, leading to the restoration of nephrin and podocin expressions. ⁷⁸ These observations have subsequently been confirmed in clinical trials. In a double‐blind, randomised, controlled trial in 40 microalbuminuric T2DM male patients, Grover‐Páez et al. demonstrated that an administration of 50 mg sildenafil once daily for 30 days to significantly reduced albuminuria (and HbA1c). ⁷⁹ In a more recent randomised, double‐blind, placebo‐controlled trial, Scheele et al. administered a long acting PDE5i once daily for 12 weeks to subjects with T2DM with overt nephropathy (eGFR 25–60 ml/min per 1.73 m²) and macroalbuminuria (urinary albumin‐to‐creatinine ratio [UACR] > 300 mg/g), in addition to a background therapy with angiotensin converting enzyme inhibitor or angiotensin receptor blocker. ⁸⁰ At 3 months, UACR reduced by 16% (95% CI, 2%–27%) compared with placebo, with values returning to baseline 4 weeks after PDE5i treatment cessation. The albuminuria‐lowering effect was not driven by systemic haemodynamic changes but possibly through the modulation of intraglomerular haemodynamics. Furthermore, lowering of HbA1c during treatment with PDE5i was observed, consistent with the previously mentioned plausible effects of PDE5 inhibition on glucose uptake in skeletal muscle, beta cell function and insulin resistance.

5. PDE5I AND WOUND HEALING

Over the past several years, NO has been proved to play an important role in the wound healing. ⁸¹ Not only does NO supports the endothelial cell proliferation but also collagen synthesis and epithelisation. ⁸² Furthermore, angiogenesis, which promotes wound tensile strength and plays a vital role in tissue repair, is dependent on the NO‐cyclic GMP axis. ⁸³ , ⁸⁴ Therefore, there has been an increased interest in the role of PDE5i in wound healing given their indirect dilatory effects on microcirculation through reduced local degradation of NO. Some preclinical studies evaluated the effects of commonly used PDE5i's on the survival of skin flaps, including vascularisation, decrease in wound size and necrosis, yielding conflicting results ⁸⁵ , ⁸⁶ , ⁸⁷ (see Table 1).

TABLE 1.

Other pleiotropic properties of phosphodiesterase 5 inhibitors (PDE5i) observed in animal studies

Study	Intervention	Outcome
PDE5i and wound healing
Kaya et al. (2015)	Acute administration of PDE5i (vs. vehicle) to rats with skin flap	The skin flap necrosis area was lower in the groups receiving oral PDE5 inhibitors than in the control group, although not statistically significant
Hart et al. (2006)	Daily (for 7 days) systemic administration of sildenafil (vs. vehicle vs. sham) to rats post‐generation of skin flap	Significant decrease in necrosis and stasis of the flap in PDE5i‐treated rats, most marked in the early post‐operative period
Souza et al. (2022)	Daily (for 14 days) systemic administration of sildenafil, or tadalafil (vs. vehicle) to rats post‐generation of skin flap	Increased vascularisation of skin flap on day 7 and 14 post‐flap generation
Abu Dayyih et al. (2020)	Daily (for 10 days) administration of sildenafil‐containing ointment, oral sildenafil, burn ointment or vehicle to skin wound in healthy and diabetic rats	The treatment with 5% sildenafil‐containing ointment associated with greater reduction in wound area in both non‐diabetic and diabetic rats, particularly in early phase of wound healing. The effect was more pronounced in non‐diabetic rats

PDE5i and diabetic neuropathy
Wang et al. (2011)	Daily administration of sildenafil (vs. vehicle) for 28 days to diabetic mice	PDE5i‐treated group demonstrated augmented axonal outgrowth and myelination, improvement in nerve conduction velocities and sensory function
Wang et al. (2015)	Daily administration of sildenafil (vs. vehicle) for 8 weeks to diabetic mice with advanced diabetic neuropathy	PDE5i‐treated group demonstrated significantly improved motor and sensory conducting velocity in the sciatic nerve, improved sensory function
Wang et al. (2016)	Tadalafil (vs. vehicle) administered every 48 h for 8 weeks to diabetic mice	PDE5i‐treated group demonstrated increased regional blood flow in the sciatic nerve tissue, improved motor and sensory conduction velocities and peripheral thermal sensitivity

In a more recent study of wound healing, topically used sildenafil was associated with a significant reduction in wound area and an increase in tensile strength when compared with treatment with oral sildenafil, negative (vehicle) or positive (burn ointment) controls in both diabetic and non‐diabetic rats. ⁸⁸ Moreover, sildenafil‐containing ointment promoted greater collagen synthesis and vascularisation in comparison with oral sildenafil‐treated rats and controls. Nonetheless, the topical PDE5i‐facilitated wound healing was much more pronounced in the non‐diabetic than in the diabetic rats. ⁸⁸ Studies looking at the effects of PDE5 inhibition on wound healing in humans are currently lacking.

6. PDE5 INHIBITOR USE IN DIABETIC NEUROPATHY

Diabetic peripheral neuropathy affects around 50% of individuals with diabetes and is associated with significant morbidity, poor quality of life and increased mortality partly resulting from the development of foot ulceration(s) and lower limb amputation. ⁸⁹ , ⁹⁰ , ⁹¹ , ⁹² , ⁹³ The clinical presentation reflects the distribution and the size of nerve fibres involved, with the distal symmetric sensory or sensorimotor neuropathy most commonly encountered.

Diabetes‐related metabolic factors result in distal axonal damage of the dorsal root ganglia neurons, which leads to numbness, pain and/or hyperaesthesia. In addition to axonal degeneration, vascular dysfunction and decreased blood flow to peripheral nerves (microangiopathy) have been described, further contributing to demyelination and degeneration of large and small nerve fibres. ⁹⁴ Conversely, a restoration of arterial blood flow was associated with an improvement in nerve conduction velocity in diabetic patients. ⁹⁵ Therefore, PDE5i have become an attractive therapeutic option for targeting neurovascular dysfunction due to their indirect effect on microvasculature.

Indeed, early preclinical studies with the use of sildenafil demonstrated improved blood supply to the vasa nervorum and functional outcome through NO ‐cGMP pathway in rodents. ⁹⁶ , ⁹⁷

Moreover, hyperglycaemia was found to be associated with the decreased proliferation of Schwann cells which play an important role in peripheral nervous system regeneration. In diabetic mice, PDE5 expression on the Schwann cells is upregulated, leading to the deactivation of cGMP‐PKG signalling pathway. Treatment of early‐stage diabetic neuropathy in mice with short‐acting PDE5i sildenafil augmented axonal outgrowth and myelination which was associated with improvement in nerve conduction velocities and sensory function. ⁹⁸ Similarly, tadalafil administered every 48 h for 8 consecutive weeks significantly improved motor and sensory conduction velocities and peripheral thermal sensitivity in diabetic mice through improvement in axon diameter and myelin thickness of the peripheral nerve. There was also a marked increase in the local blood flow with the subsequent increase in intraepidermal nerve fibre density. ⁹⁹

As far as the treatment of advanced diabetic peripheral neuropathy with PDE5i is concerned, the preclinical studies suggest possible benefits in terms of enhanced regional blood flow, increased myelin sheath thickness and improved motor and sensory conduction velocities as well as thermal sensitivity. However, a significant increase in the axonal diameter was not observed, and the overall clinical improvement in sensory function was not evident after a longer period of treatment. ¹⁰⁰ The functional benefits are likely a result of a combination of the above‐mentioned cellular and histological changes, and it is possible that the PDE5i enhance neurovascular function through multiple signalling pathways. The early intervention appears to have the best possible outcome.

The evidence in humans is largely limited to case reports which indicated an improvement in neuropathic symptoms in patients with diabetes receiving PDE5i regularly ¹⁰¹ ; however, large‐scale clinical trials are needed to determine if PDE5i would indeed be of benefit to patients with diabetic peripheral neuropathy.

7. PDE5 INHIBITORS AND TESTOSTERONE LEVELS IN MALE SUBJECTS

Yet another aspect of PDE5 inhibition which has been investigated in the recent years is the effect of PDE5i on Leydig cell steroidogenesis following an observation of an increase in total testosterone levels with prolonged PDE5i treatment. ¹⁰² , ¹⁰³ However, such studies are limited and conducted predominantly in non‐selected populations where generalisability to diabetic cohort might be limited.

In a recent animal study, 12 weeks of tadalafil administration to diabetic and non‐diabetic rats resulted in significant increase in total and free testosterone levels in both groups. ¹⁰⁴ In a double‐blind, randomised controlled trial in 54 men with T2DM and erectile dysfunction, 24‐week treatment with twice daily vardenafil not only improved the erectile and systemic endothelial function but also in a subgroup of 7 men with hypogonadism (total testosterone <10.4 nmol/L), a significant increase in total testosterone level into the eugonadal range was observed after 1 week of PDE5i administration. ¹⁰⁵ The effect persisted throughout the treatment phase; total testosterone levels rapidly decreased to previous hypogonadal values on the cessation of PDE5i. Interestingly, no concurrent changes in luteinising hormone levels were observed. Among eugonadal men, total testosterone levels did not significantly change throughout the study period. ¹⁰⁶

In unselected human cohort, an earlier retrospective open‐label study demonstrated that the on‐demand use of sildenafil or tadalafil for 3 months by men with erectile dysfunction was associated with a statistically significant increase in total and free testosterone as well as a decrease in luteinising hormone. ¹⁰⁷ These changes, however, occurred within the normal reference range for both hormones. Although it has been postulated that the observed increase in total testosterone level could be due to the increase in sexual activity with PDE5i use, more recent preclinical evidence suggests that prolonged PDE5 inhibition might directly stimulate Leydig call steroidogenesis via cAMP/cGMP signalling pathway. ¹⁰³ Greater sexual activity did not predict greater change in serum total testosterone in the testosterone and erectile dysfunction trial where an optimised dose of sildenafil was administered to 140 men, aged 40–70 years, with erectile dysfunction, low serum total testosterone (<11.4 nmol/L) and/or free testosterone (<173 pmol/L) over 3–7 weeks. ¹⁰⁸ Statistically significant mean increases of 3.6 and 110 pmol/L in total and free testosterone levels, respectively, were observed. This was accompanied by parallel increases in serum dihydrotestosterone, oestradiol and significant suppression of luteinising hormone level, again suggesting a direct effect of PDE5i at the testicular level. ¹⁰⁸

Conversely, Jarvi et al. and Hellstrom et al. demonstrated that neither sildenafil, tadalafil nor vardenafil taken daily for 6 months significantly changed total testosterone and/or gonadotrophin levels in eugonadal men. ¹⁰⁹ , ¹¹⁰

The research to date has been conflicting yet, there is a suggestion that the increase in testosterone level with PDE5 inhibition might be more specific to the hypogonadal individuals. These findings warrant further research in this specific cohort.

8. CONCLUSION

The preclinical research data largely support the notion that the PDE5 inhibition might result in a number of favourable systemic effects in individuals at risk of or with established cardiovascular disease and/or metabolic dysfunction. The restoration of an NO‐cGMP‐PKG signalling pathway through PDE5 inhibition appears to decrease oxidative stress and augment some of the micro‐ and macrovascular complications of diabetes. Nonetheless, the clinical data are insufficient and largely limited by small study sample size and heterogeneity of the studied cohorts (Table 2). Moreover, very few studies specifically focus on diabetes population. Adequately powered, randomised‐controlled trials in diabetic cohorts with varying duration of diabetes and related complications are needed to fully assess the benefits of PDE5i in this group of patients.

TABLE 2.

Pleiotropic effects of phosphodiesterase type 5 (PDE5) inhibitors in human studies

Study	Design	Participants	Intervention	Outcome
PDE5i use in cardiovascular disease
Guazzi et al. (2011)	Double‐blind, randomised, placebo‐controlled trial	A total of 45 heart failure (LVEF < 40%, NYHA II‐III) non‐diabetic patients	Sildenafil 150 mg daily or placebo for 1 year	Sildenafil group at 6 months and 1 year: increase in LV ejection fraction, diastolic function, improved exercise performance and quality of life
Lewis et al. (2007)	Placebo‐controlled, double‐blind, parallel‐group, single‐centre study	A total of 34 subjects with LVSD (NYHA II‐IV) and secondary PH	Sildenafil 25–75 mg three times a day or placebo for 12 weeks	Sildenafil group: improved exercise capacity (reduced pulmonary vascular resistance and increased cardiac output with exercise) and 6‐min walk distance, improved quality of life
Guazzi et al. (2011)	Double‐blind, randomised, placebo‐controlled trial	A total of 44 subjects with HF and preserved ejection fraction (LVEF ≥50%)	Sildenafil 50 mg three times a day or placebo for 6 months	Sildenafil‐treated group at 6 and 12 months: improvements in mean pulmonary artery pressure, RV function and lung interstitial water metabolism
Gianetta et al. (2012)	Double‐blind, randomised, placebo‐controlled trial	A total of 59 men with non‐ischaemic diabetic cardiomyopathy	Sildenafil 100 mg a day or placebo for 12 weeks	Sildenafil treatment associated with anti‐remodelling effect, resulting in improved cardiac kinetics. No changes were found in endothelial function
Redfield et al. (2013)	Multicentre, double‐blind, placebo‐controlled, parallel‐group, randomised trial (RELAX)	A total of 216 subjects (median age 69) with HF and preserved ejection fraction (LVEF ≥ 50%), 40% had underlying diabetes	Sildenafil 20 mg three times daily for 12 weeks, followed by 60 mg three times daily for 12 weeks, or placebo	At 24 weeks, no change in exercise capacity and clinical status including QoL between the two study arms
Thadani et al. (2002)	Double‐blind, crossover, single‐dose multicentre study	A total of 41 men (mean age 62) with reproducible stable exertional angina due to ischemic CAD	A volume of 10 mg vardenafil or placebo given 1 h prior to treadmill exercise	Prolonged time to ischemic threshold following vardenafil administration relative to placebo
Fox et al. (2003)	Double‐blind placebo‐controlled trial	A total of 108 men (mean age 63) with evidence of CAD	A volume of 100 mg sildenafil or placebo given 1 h prior to treadmill exercise	Sildenafil demonstrated superiority to placebo for the time to onset of angina, the time to limiting angina and the total exercise time
Denardo et al. (2011)	Open‐label, non‐randomised prospective cohort design	A total of 23 women (mean age 54) with microvascular coronary dysfunction and without macrovascular obstructive coronary artery disease	A volume of 100 mg sildenafil given 45 min prior to angiography	Acute improvement in coronary flow reserve post‐PDE5i
Dietz et al. (2003)	Open‐label, non‐randomised prospective cohort design	A total of 21 men (mean age 60) with diabetes and ED	A volume of 50 mg sildenafil given 1 h prior to echocardiography	No change in coronary flow reserve post PDE5i
Anderson et al. (2016)	Observational cohort study	A total of 5956 men aged 40–89 with T2DM	–	Use of PDE5i (n = 1359) was associated with lower risk of overall mortality and mortality in those with a history of acute MI
Gazzaruso et al. (2008)	Observational cohort study, median follow‐up 47 months	A total of 291 men (mean age 55) with T2DM and silent CAD	–	PDE5i and statin use was associated with lower incidence of major adverse cardiac events during follow‐up period
PDE5i and glucose metabolism
Gonzalez‐Ortiz et al. (2007)	Randomised, double‐blind, placebo‐controlled clinical trial	A total of 18 obese men aged 30–50 without diabetes	Tadalafil 5 mg a day or placebo for 4 weeks	No difference in insulin secretion or insulin sensitivity between the two groups
Ramirez et al. (2015)	Randomised, double‐blind, placebo‐controlled study	A total of 42 overweight subjects (mean age 51) with pre‐diabetes	Sildenafil 75 mg a day or placebo for 3 months	Sildenafil group: improved insulin sensitivity and markers of endothelial function. Improved urinary ACR
Ho et al. (2014)	Randomised, double‐blind, placebo‐controlled study	A total of 53 obese subjects (mean age 33) with elevated fasting insulin levels	Tadalafil 20 mg a day or placebo for 3 months	Improvement in insulin resistance only in severely obese subjects (BMI ≥ 36.2 kg/m2)
Jansso et al. (2010)	Randomised, placebo‐controlled, crossover study	A total of 17 postmenopausal women with T2DM	A single dose of tadalafil 20 mg or placebo	Increased muscle glucose uptake after PDE5i administration
Sjogren et al. (2016)	Randomised, placebo‐controlled study	A total of 20 subjects with T2DM	A single dose of tadalafil 20 mg or placebo	Increased muscle glucose uptake after PDE5i administration
PDE5i and diabetic nephropathy
Grover‐Páez et al. (2007)	Double‐blind, randomised, controlled trial	A total of 40 males with T2DM and microalbuminuria	A volume of 50 mg sildenafil once daily for 30 days	Overall, 51% reduction in albuminuria and 0.6% lower HbA1c
Scheele et al. (2016)	Randomised, double‐blind, placebo‐controlled trial	A total of 256 men and women with eGFR 25‐60 ml/min and macroalbuminuria, on ACEi or ARB background treatment	Long acting PDE5i PF‐00489791 (20 mg) or placebo orally, once daily for 12 weeks	Overall, 16% reduction in urinary albumin‐to‐creatinine ratio; 0.3% lower HbA1c
PDE5i's and their effect on reproductive hormone levels
Santi et al. (2016, 2017)	Randomised, placebo‐controlled, double‐blind clinical trial	A total of 54 male patients with T2DM and ED	Vardenafil 10 mg twice a day or placebo for 24 weeks	Improvement in erectile function and flow‐mediated dilation of brachial artery; increase in total testosterone level only among men with baseline low TT (<10.4 nmol/L)
Carosa et al. (2004)	Open‐label, retrospective study	A total of 74 men eugonadal with ED	On demand sildenafil 50 mg or tadalafil 20 mg for 3 months	Significant increase in TT, free T and decrease in LH
Spitzer et al. (2014)	Randomised, double‐blind, placebo‐controlled, single‐centre, and parallel group trial	A total of 140 men aged 40–70 with ED and TT (< 11.4 nmol/L) and/or free T (<173 pmol/L)	Optimised dose of sildenafil for 3–7 weeks	Sildenafil use associated with significant increases in TT, free T, DHT and E2 and suppression of LH levels
Jarvi et al. (2008)	Randomised, double‐blind, placebo‐controlled, parallel group study	A total of 200 eugonadal men aged 25–64, with and without ED	A volume of 20 mg vardenafil or 100 mg sildenafil or placebo taken daily for 6 months	Reproductive hormone levels unaffected by the use of either of the two PDE5 inhibitors
Hellstrom et al.	Randomised, double‐blind, placebo‐controlled study	A total of 421 men >45 years old with or without erectile dysfunction	A volume of 10 or 20 mg tadalafil or placebo taken daily for 6 months	Reproductive hormone levels unaffected by the PDE5 inhibitor use

Abbreviations: ACEi, angiotensin‐converting enzyme inhibitor; ACR, albumin–creatinine ratio; ARB, angiotensin II receptor blocker; BMI, body mass index; CAD, coronary artery disease; DHT, dihydrotestosterone; E2, oestradiol; ED, erectile dysfunction; eGFR, estimated glomerular filtration rate; free T, free testosterone; HF, heart failure; LV, left ventricle; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association; PH, pulmonary hypertension; QoL, quality of life; RV, right ventricle; T2DM, type 2 diabetes mellitus.; TT, total testosterone.

Swiecicka A. The efficacy of PDE5 inhibitors in diabetic patients. Andrology. 2023;11:245–256. 10.1111/andr.13328

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analysed in this study.