This prospective, observational, cohort study assessed the following age-matched patient groups: 33 patients with newly diagnosed AC (mean age 47 years; 18 males; 54.5% of patients with controlled AH – the mean blood pressure [BP] of 121/77 mmHg), with no other significant comorbidities and 155 patients from the control group (CG) (mean age 47 years; 106 males; 77.4% of patients with controlled AH – the mean BP 119/76 mmHg) with no significant comorbidities.

The study was conducted in accordance with the Declaration of Helsinki and the principles of good clinical practice (GCP). The study protocol had been approved by the institutional ethics committee at the Military Institute of Medicine in Warsaw (approval No. 76/WIM/2016). A written informed consent had been obtained from each patient.

The AC group comprised newly diagnosed patients with clinically active acromegaly, defined based on the standard endocrine and imaging criteria in accordance with European Society of Endocrinology (ESE) guidelines: co-occurrence of the characteristic somatic features of AC and laboratory test abnormalities (the GH and IGF-1 levels above the upper limits of normal for the sex and age and a lack of GH suppression to less than 46 pmol/L (1.0 mcg/L) by a glucose load of 75 g (oral glucose tolerance test, OGTT), and a focal pituitary lesion visualized via magnetic resonance imaging (MRI) (20). All AC group patients underwent pituitary hormone assessments (adrenocorticotropic hormone [ACTH], follicle-stimulating hormone [FSH], luteinizing hormone [LH], thyroid stimulating hormone [TSH]) and triiodothyronine (FT3), thyroxine (FT4), cortisol and total testosterone and any preexisting or newly diagnosed carbohydrate metabolism impairment was recorded; this included type 2 diabetes mellitus (T2DM), impaired fasting glycemia (IFG), and impaired glucose tolerance (ITG). The criteria for the diagnosis of other comorbidities were applied according to the latest guidelines (21, 22).

The CG patients were the population taking part in the State-funded study “Non-invasive haemodynamic assessment in hypertension (FINE-PATH)” NCT01996085 (ClinicalTrials.gov) conducted at the Military Institute of Medicine. From this group, smaller subgroups, with comparable proportions of AH patients, were selected and matched for key clinical parameters [age, sex, body mass index (BMI), mean blood pressure (MBP)] for the purpose of each comparative analysis. The CG included 120 patients of either sex, with AH treated for at least 12 months and 35 healthy individuals of either sex, with no cardiovascular or other significant internal diseases, who had provided their consent to take part in this study.

Study exclusion criteria were: coronary artery disease, including a history of myocardial infarction; chronic HF with the LVEF of <50%; history of pulmonary embolism, history of a documented stroke or transient ischemic attack (TIA); chronic obstructive pulmonary disease (COPD); respiratory failure (arterial partial pressure of oxygen [PaO₂] of < 60mmHg and/or the partial pressure of carbon dioxide [PaCO₂] > 45mmHg); history of head injury; pregnancy; a lack of consent.

The clinical assessment focused on cardiovascular risk factors, including a family history of cardiovascular conditions; cardiovascular symptoms; comorbidities; nicotine dependence; impaired carbohydrate metabolism; lifestyle; office blood pressure measurement (OBPM) (systolic and diastolic BP [SBP and DBP]), heart rate (HR), and BMI. The standard OBPMs were conducted with an automatic sphygmomanometer (Omron M4 Plus, Japan) in accordance with European Society of Cardiology (ESC) guidelines (21).

All patients with AC underwent hemodynamic parameter assessment via ICG with a Niccomo™ device (Medis, Ilmenau, Germany). This study analysis was performed based on the ICG recordings obtained during 10-minute assessments of the following parameters measured at rest in a horizontal position: HR; SBP; DBP; cardiac pump function: cardiac output (CO [mL/min]), cardiac index (CI [L/min/m²]), stroke volume (SV [mL]), stroke volume index (SI [mL/m²]); cardiac contractility parameters: velocity index (VI [1/1000/s]: VI=1000*dZmax*Z0⁻¹), reflecting the peak aortic blood flow velocity; acceleration index (ACI [1/100/s²]: ACI=100*dZmax*dt⁻¹), reflecting the peak aortic blood acceleration; and Heather index (HI [Ohm/s²]: HI=dZmax*TRC), reflecting cardiac contractility. Moreover, systemic vascular resistance (SVR [dyn*s*cm⁻⁵], systemic vascular resistance index (SVRI [dyn*s*cm⁻⁵*m²]), and thoracic fluid content (TFC [1/kOhm]) were also measured. Consistent with the PREDICT study, which helped identify different risk groups based on the SI and TFC values, the cutoff values adopted in this study were < 35 mL/m² and > 35 1/kOhm, respectively (23).

Digital filing and statistical analysis of study data were conducted with MS Office and Statistica 12.0 (StatSoft Inc., Tulsa, Oklahoma, US). Continuous variables were presented as means ± standard deviation (SD), medians, and interquartile ranges; and categorical (qualitative) variables were presented as absolute values (n) and proportions (%). Continuous variable distribution was assessed visually and with the Shapiro–Wilk test. For each comparative analysis, subgroups comprising similar proportions of AH patients were selected from the CG by propensity score matching of the key clinical parameters (age, sex, BMI, MBP) that might noticeably affect the evaluated parameters. The differences between absolute values of continuous variables were analyzed with the t-test for normally distributed variables and the Mann–Whiney U test for non-normally distributed data. Categorical (qualitative) variables were analyzed with the chi-square and Fisher tests. Statistical significance was set at p < 0.05.

All the baseline characteristics of AC and CG patients have been presented in Table 1 . Eighteen patients (54.5%) with AC were also diagnosed with AH. In all cases, the AH had been treated with medication, usually with one or two antihypertensive agents. Out of the patients with both AC and AH, twelve patients were treated with ACEI, one patient with ARB, eight patients with CCB, six patients with beta-blockers and four patients with diuretics. T2DM was diagnosed in 6 (18.2%) of the 33 patients with AC, and prediabetes was diagnosed in 10 patients (30.3%), whereas 17 patients (51.5%) exhibited normal glucose tolerance. Out of the patients with both AC and T2DM, two patients were treated with metformin and three with metformin and insulin. Thirty-one out of the 33 patients with AC had a functioning anterior pituitary lobe. Two patients with an invasive somatotropic tumor were diagnosed with TSH deficiency; however, this was well controlled with a stable L-thyroxin dose.

In the CG, 120 patients (77%) had AH. In all these patients, AH had been treated with medication for at least 12 months.

The evaluated ICG parameters and their values measured in the AC group have been presented in Table 2 .

A comparative analysis of the baseline characteristics in the AC group and in the CG has been presented in Table 3 .

The analysis revealed significant differences in the HR and serum creatinine levels. Patients with AC were characterized by significantly higher HR values and significantly lower serum creatinine levels than those in the CG (HR of 76.2 bpm vs. 66.8 bpm, respectively; p=0.001; and creatinine of 0.76 mg/dL vs. 0.85 mg/dL, respectively; p=0.035). No significant differences between the groups were observed in any of the other evaluated parameters. A comparative analysis of the evaluated ICG parameters showed a number of differences between the groups ( Table 4 and Figure 1 ). In comparison with CG patients, the patients with AC were shown to have significantly higher HR values (p=0.003), significantly lower SI (p=0.0001), significantly lower indices of cardiac contractility: VI (p=0.037) and HI (p<0.0001), and significantly higher TFC (p<0.0001).

A comparative analysis of the CG and AC patients in terms of other selected parameters has been presented in Table 5 . ICG assessments showed a low SI (< 35 mL/m²), which reflects impaired LV function, in 3.1% of CG patients and in 18.8% of AC patients (p=0.045). ICG assessments also showed a high TFC (> 35 1/kOhm) in 3.1% of patients from the CG and in 68.8% of patients from the AC group (p< 0.0001).

There was no significant difference between AC patients with hypertension and those without hypertension for any of the evaluated hemodynamic parameters. There was also no significant difference between AC patients with and without diabetes or glucose intolerance for any of the evaluated hemodynamic parameters. No significant correlations between hemodynamic parameters and GH/IGF1 were noted ( Supplementary Table S1 ).

The development of cardiovascular complications in patients with AC is responsible for an increases mortality due to structural and functional alterations in the cardiovascular system (36, 38). One of the most common adverse consequences of long-term tissue exposure to GH excess in patients with AC is the development of AH. It has been already shown that an altered circadian BP profile in patients with active acromegaly (non-dipping profile) may be associated with higher GH and IGF-1 levels (39). In our study, 54.5% out of the 33 patients with AC were diagnosed with AH. Importantly, AH is also an independent predictor of mortality in AC patients, with the following factors potentially contributing to the development of AH in this patient group: rapid HR, increased LV afterload, increased SV and CO, endothelial dysfunction, changes in systemic vascular resistance, and reduced coronary perfusion pressure (24, 40), with direct antidiuretic effects of GH and IGF-1 in the kidneys also playing an important role (41). Excessively high levels of these hormones may increase the renal reabsorption of sodium and secondary reabsorption of water, which leads to an increased plasma volume and vascular response to angiotensin, and consequently an increased systemic vascular resistance (7). In our study, AC patients showed a tendency to have high TFC values, which suggests an important role of high TFC in the process of AH development in this group of patients; however, there are a number of authors who emphasize that a high TFC does not seem to be the key factor in AH development in patients with AC (20). The tendency for fluid retention can be explained as a pathophysiologic phenomenon. Both GH and IGF-1 have antidiuretic effects, which lead to increased plasma volume. If this coincides with other GH- and IGF-1-induced problems (such as overstimulation of the smooth muscle tissue in small blood vessels and the resulting fibrosis, vasoconstriction, insulin resistance, hyperinsulinemia, and excessive sympathetic system activation), the risk of inadequately controlled AH increases considerably (7, 38, 41). Moreover, an increased TFC may be also responsible for subclinical myocardial dysfunction, which seems to be indicated by the results described below.

Our study showed that patients from the AC group have lower cardiac pump function parameters (SI, HI, and ACI) than those in the CG. These findings were also confirmed via echocardiographic examinations, with AC patients exhibiting lower LVEF values than CG patients. These findings are consistent with those reported by other authors. Moreover, many study reports emphasize that many patients with AC exhibit structural myocardial remodeling, which is associated with LV hypertrophy and concentric remodeling (14, 24). This is followed by myocardial fibrosis, which is directly associated with GH effects, not only with LV hypertrophy or AH. Cardiac fibrosis impairs LV hemodynamic function, which first manifests in the form of a diastolic dysfunction and later leads to impaired systolic function and the development of symptomatic HF (2, 8, 9, 25, 42, 43). Diastolic LV dysfunction affects approximately 30% of untreated AC patients (41). Importantly, over 60% of adults with AC suffer from AH, and the length of time from AC diagnosis seems to be correlated with AH development. AH, which affects over 60% patients with AC, significantly exacerbates the structural and functional LV dysfunction. One study demonstrated that it was DBP that was the best predictor of LV hypertrophy (8). The observed increases in both CO and CI values during early AC-associated cardiomyopathy and the reduction in SVRI values have also been considered in a search for the possible mechanisms behind AH development in AC (8, 9, 13, 44). Our study showed low values of cardiac pump function parameters (SI and CI) in AC patients, even despite well-controlled AH; this corresponded with echocardiographic evidence of abnormal LV function. Studies in other populations of patients with HF and essential AH have demonstrated a correlation between the values of ICG-assessed hemodynamic parameters and those of echocardiographic indices of LV function (17–19, 45, 46).

The results of our study also demonstrated that AC patients were characterized by a significantly higher HR and a slightly higher CI in comparison with those in the CG. These findings are consistent with earlier reports of concentric LV hypertrophy and increased contractility and HR during this early, subclinical, and reversible phase of cardiomyopathy (2, 8, 10). Therefore, hemodynamic abnormalities initially manifest as a hyperkinetic heart, and not until later do they take the form of impaired cardiac pump function (8, 13, 47). Lower creatinine levels in acromegalic patients as compared to controls could be partially explained by renal hypertrophy and consequently enhanced renal filtration process (48).

The study presented here showed that ICG results may be an important indicator of subclinical LV dysfunction in AC patients, even in those with well-controlled AH. Due to its capacity to assess the cardiovascular hemodynamic function in AC patients, ICG seems to be of considerable clinical importance, which is measured via the usefulness of this technique in risk stratification in this patient group. It seems particularly important to detect hemodynamic abnormalities in this population before clinical symptoms develop, i.e. at the stage of subclinical disease, when an early therapeutic intervention may prevent fully symptomatic cardiovascular complications. The pathological mechanisms of AH in this patient population are a good argument for initiating antihypertensive combination therapy of vasodilators (angiotensin-converting enzyme [ACE] inhibitors; calcium channel blockers [CCB]) and diuretics.

The main limitation of the study was the relatively small sample size. This was due to a low prevalence of AC due to a hormonally active pituitary tumor. Many patients with AC had already exhibited comorbidities or evidence of severe cardiovascular dysfunction at the time of diagnosis, and were thus excluded from this study, further diminishing the size of the study population. ICG result interpretation must account for the potential effects of the antihypertensive therapy used i.e. cardiodepressive action of beta-blockers, vascular effect of ACEI/CCB and influence of diuretics on fluid status. However, no significant intergroup differences in pharmacotherapy limits these bias in our study. Likewise, the effect of patient sex on hemodynamic abnormalities in hormone secreting pituitary tumors requires further studies.