Serum pro-brain natriuretic peptide correlates with optical coherence tomography indices in diabetic retinopathy

• Introduction
• Methods
• Results
• Discussion

Abstract

Purpose: Serum pro-brain natriuretic peptide (BNP) is a 108-amino-acid prohormone that inhibits vascular endothelial growth factor (VEGF) secretion, protecting pericytes from cell death and decreasing retinal vascularization. The purpose of this study was to investigate the correlation of serum pro-BNP with optical coherence tomography (OCT) indices in diabetic retinopathy.

Methods: This cross-sectional study investigated 96 consecutive subjects aged between 40 and 65 years: controls n = 24, no diabetic retinopathy (NoDR) n = 24, non-proliferative diabetic retinopathy (NPDR) n = 24, and proliferative diabetic retinopathy (PDR) n = 24. Same-day analysis of blood samples for serum pro-BNP levels was performed and spectral-domain OCT (SD-OCT) was used to measure the following OCT indices: OCT angiography (OCTA) superficial vessel density (SVD), deep vessel density (DVD), and foveal avascular zone (FAZ); OCT retinal nerve fiber layer (RNFL); and OCT ganglion cell analysis (GCA).

Results: The mean serum pro-BNP levels for the control, NoDR, NPDR, and PDR groups were 14.07 ± 11.51, 27.35 ± 11.81, 280.44 ± 106.13, and 122.33 ± 43.66 pg/ml, respectively. The mean values of the various OCT parameters correlated with serum pro-BNP were OCTA SVD (r = - 0.360), OCTA DVD (r = 0.408), OCTA FAZ (r = 0.475), OCT RNFL (r = - 0.215) and OCT GCA (r = - 0.285; p<0.001).

Conclusions: The serum pro-BNP levels were higher in the NPDR group than in the NoDR group and much lower in the PDR group than in the NPDR group, reflecting a lowering of the protective barrier. These results correlated with the changes in various OCT indices.

A sight-threatening microvascular consequence of diabetes mellitus (DM) is diabetic retinopathy (DR) [1]. A strong correlation exists between the duration of diabetes and the prevalence of DR, with the incidence rising from 28.8% in individuals with diabetes for less than 5 years to 77.8% in those with diabetes for over 15 years [2].

Optical coherence tomography (OCT) is used to perform an optical biopsy of the retina [3] for the assessment of retinal diseases, while OCT angiography (OCTA) distinguishes the fine microvascular features of the retinal vasculature in the superficial and deep retinal plexus without the need for fluorescein dye [4,5].

Pre-pro-brain natriuretic peptide (BNP), a 134-amino-acid precursor secreted by cardiac myocytes in response to volume overload, is cleaved into a signal peptide (26 amino acids) and a 108-amino-acid prohormone pro-BNP [6]. Following its release into the bloodstream, pro-BNP is broken down into the physiologically active hormone BNP (77–108 amino acids) and the inactive metabolite N-terminal fragment of the pro-brain natriuretic peptide (NT pro-BNP; 1–66 amino acids) by the proteases furin and corin [7-9]. Because pro-BNP is filtered by the kidneys, unlike BNP, which is digested in the systemic circulation, plasma pro-BNP concentrations are higher than BNP concentrations. With a normal glomerular filtration rate, pro-BNP has a longer half-life and a more stable in vitro chemical structure than BNP [10], leading to slower oscillations and higher circulating levels [11]. Therefore, after blood collection, serum levels of pro-BNP remain steady, but those of BNP fluctuate. Thus, pro-BNP levels in plasma may serve as a reliable indicator of disease. In participants free of clinical cardiovascular disease, higher levels of NT pro-BNP are associated with retinal microvascular damage, suggesting a potential role for NT pro-BNP as a marker for small vessel disease [12].

The understanding, monitoring, and management of DR could be facilitated by investigating the correlation between serum pro-BNP levels and various OCT indices in DR. Hence, a tertiary care center-based study was undertaken for the first time.

A total of 72 consecutive patients with type 2 DM and 24 healthy controls between the ages of 40 and 65 years presenting to this tertiary care center were included. The patients were categorized into three groups based on the Early Treatment Diabetic Retinopathy Study (ETDRS) classification: no diabetic retinopathy (NoDR, n = 24), non-proliferative diabetic retinopathy (NPDR, n = 24), and proliferative diabetic retinopathy (PDR, n = 24) [13].

Approval from the institutional ethics committee was granted. The study was registered with the Trial Registry of India (ECR/262/Inst/UP/2013/RR-19) and followed the tenets of the Declaration of Helsinki. Informed consent was obtained from the subjects after an explanation of the nature of the study.

The inclusion criteria were eyes with natural crystalline lenses and eyes with nuclear sclerosis up to grade 2 (LOCS III) [14]. In patients with bilateral involvement, the eye with more severe disease was included, and patients with OCT scans of signal strength exceeding 5 were included. Individuals who had previously undergone laser treatment, ocular surgery, or intravitreal injections were excluded. Furthermore, people with systemic disorders that could influence the retinal vasculature, such as chronic kidney disease, pulmonary diseases (e.g., pulmonary embolism and pulmonary hypertension), chronic obstructive pulmonary disease, essential hypertension, or cardiovascular diseases (e.g., heart failure with reduced or preserved ejection fraction, right ventricular failure, valvular heart disease, coronary artery disease, and myocardial diseases, such as left ventricular hypertrophy and myocarditis, and arrythmias) were also excluded.

The comprehensive ophthalmological workup included best corrected visual acuity assessment (logMAR BCVA), slit lamp biomicroscopy, and fundus evaluation with a +90D lens and indirect ophthalmoscopy. The diagnostic workup included OCTA superficial vessel density (SVD), deep vessel density (DVD), and foveal avascular zone (FAZ), and the OCT retinal nerve fiber layer (RNFL), and OCT ganglion cell analysis (GCA) were performed using SD-OCT (Cirrus HD-OCT, Carl Zeiss Meditech, Inc., Dublin, CA).

The biochemical workup included collection of whole blood samples of 6–7 ml using a metal-free plastic syringe and a 24-gauge stainless steel needle for aseptic venipuncture and same-day analysis of blood samples. On an autoanalyzer, the conventional technique was used (automatic biochemical analyzer) for measuring fasting blood glucose (FBS) and post-prandial blood glucose (PPBS). For the glycosylated hemoglobin (HbA1c) level measurement, standard methods (mass spectroscopy and chromatographic techniques) were used. Serum pro-BNP levels were estimated on a Roche Elecsys cobas e 411 (Roche Diagnostics Limited, Basel, Switzerland) based on the test principle of electrochemiluminescence. The total duration of the assay was 18 min.

Table 1 shows the association of various demographic, clinical, and biochemical variables with the severity of DR. Of the 96 study subjects, 54 were male and 42 were female. No significant differences were observed in age and sex between the cases and controls (p>0.05). Various DR stages showed a statistically significant association with FBS, PPBS, HbA1c, and serum pro-BNP (p<0.001).

Figure 1 depicts the distribution of serum pro-BNP according to DR severity. Table 2 shows statistically significant correlations between different OCT indices and DR severity. The mean SVD and DVD were observed to decrease along with the RNFL and ganglion cell thinning. The mean FAZ increased with DR severity (p<0.001).

The values of various OCT indices associated with different DR stages are presented in Figure 2, Figure 3, Figure 4, Figure 5, and Figure 6. The bivariate analysis showed that serum pro-BNP had a negative correlation with the OCT indices, whereas OCTA FAZ had a positive correlation (p<0.01; Table 3). Table 4 presents a comparison of the means for all variables across the study groups with the control group as well as the comparisons between the study groups. Similarly, Table 5 shows the comparison between the means of all the values of the various OCT indices in the study groups and the control group, as well as with each other.

Our study’s purpose was to gain an understanding of the relationship between OCT parameters, serum pro-BNP, and the presence of DR. New insights into the processes underlying natriuretic peptide (NP) have contributed to the acknowledgment of these peptide hormones as a significant system governing cardiovascular function, as well as the regulation of water and electrolyte balance. These substances dilate the blood arteries and intensify natriuresis and diuresis. In clinical practice, BNP, a neurohormone, is widely employed as a diagnostic and predictive biomarker for heart failure [15]. It is becoming evident that BNP can mediate peripheral vasodilation and affect fibrosis, inflammation, and oxidative stress [16]. Changes in circulating BNP levels have also been reported to contribute to diabetic microangiopathies, including diabetic nephropathy and DR [17]. Our recent study highlighted serum pro-BNP as a novel molecular biomarker for PDR [18].

The retinal microvasculature is critical for the visual function of the neural retina. In the retinal microvasculature, pericytes and endothelial cells are the two key cellular elements. Pericyte–endothelial interactions are crucial for the integrity and functionality of the retinal neurovascular unit, including vascular cells, retinal neurons, and glial cells. The equilibrium of the microvasculature requires pericyte–endothelial interactions, which are disturbed in DR. Oxidative stress and inflammation in DR can disturb the critical communication between pericytes and endothelial cells, thereby resulting in the breakdown of the blood–retinal barrier [19]. Pro-BNP and concurrent microvascular issues have been found to be significantly associated with DR [20]. Higher levels of pro-BNP have been linked to retinal microvascular damage in individuals without clinical cardiovascular illness, indicating a possible role for pro-BNP as a marker for small vessel disease [21]. The inner and outer plexiform layers of the rat retina have been found to contain atrial natriuretic peptide [22]. Additionally, transforming growth factor-β1 (TGF-β1)-induced pericyte loss has been found to be inhibited by NP signaling [23]. Improving endogenous NP/NPR-A/cGMP signaling could lead to new treatment options for retinopathies associated with neovascularization by identifying a new target for medication.

In this study, mean serum pro-BNP levels were higher in the NPDR group than in the No-DR group. The levels in the PDR group were much lower than in the NPDR group. This can be explained by the fact that BNP inhibits VEGF secretion via TGF-β1 and cyclic guanosine monophosphate signaling in pericytes and astrocytes, thereby protecting pericytes from death and decreasing retinal vascularization. VEGF suppresses NP secretion from endothelial cells, while NP reduces VEGF production [23,24]. When the proliferative DR stage occurs, BNP levels decrease, which lowers the defense mechanisms and disturbs this equilibrium [18].

Hee et al. [25] reported that OCT may be used to both objectively and subjectively track retinal thickness over time due to its high repeatability. Our study found that all OCT variables, representing neovascularization, played a significant role in the transition from the NoDR to NPDR to PDR stages. Circulating levels of pro-BNP were significantly increased in patients with DR. Retinal vascularity was reduced when blood serum pro-BNP levels were increased, as shown by a decrease in the SVD and DVD values and an increase in the FAZ mean values on OCTA. An increase in serum pro-BNP was also associated with damage to the cell body (reflected in the ganglion cell layer) and axons (reflected by the RNFL), as seen from the decreasing mean OCT RNFL and OCT GCA values.

The present study has a few limitations. It was a cross-sectional study with a small sample size. Another limitation is the lack of measurement of pro-BNP levels in the vitreous in our study groups. To the best of our knowledge, no such studies have been reported in the available literature. Investigating serum pro-BNP levels in the vitreous is a potential area for exploration in our future interventional studies, particularly for the PDR group. Since this was an observational study, no interventions were performed in the study groups. However, for the first time, this study highlighted the extent of the association of serum pro-BNP with various OCT parameters in DR to facilitate the earlier detection and management of retinal changes. It can be concluded that serum pro-BNP levels increase with increasing severity of DR and correlate well with various OCT indices.