Discussion
COPD patients have increased UACR compared to controls which was related to the increased aortic stiffness. The UACR increase in the presence of similar biomarkers of proximal tubular damage would reinforce the glomerular damage leading to increased permeability in patients. Aortic stiffness was an independent variable of UACR in the setting of similar peripheral BP measures, with BP if anything marginally lower in patients with COPD.
The recent GOLD document has embedded that comorbidities are important to detect and manage in patients with COPD. One of the co-morbidities highlighted as frequent and important was cardiovascular disease. Recommendations encourage healthcare professionals to manage the co-morbidity in general, as if the patient does not have COPD. However, this fundamentally relies on diagnosing the co-morbidity in the first instance and a robust platform of assessment to monitor response and progression. If renal changes are related to the altered haemodynamics, it is imperative that we fully evaluate the interplay of microvasculature to the macrovascular state in COPD. In COPD patients, associations of aortic stiffness with UACR were demonstrated despite only a small proportion above the threshold defining clinical microalbuminuria. The striking increased aortic PWV in those patients with newly identified microalbuminuria reinforces the clinical context. Albuminuria, though, is a continuum and measureable renal injury occurs below the arbitrary threshold for microalbuminuria.
A decision was made to include all patients with confirmed COPD, irrespective of co-existent comorbidities such as IHD and diabetes. This is a different approach to previous studies we have conducted but has been widely adopted by others. We opted a priori that this would a) represent a more typical COPD population seen in clinic and b) if there are subclinical changes in patients, then exclusion of a subset with prior diagnosed disease seemed arbitrary and may relate to diagnosis as opposed to presence.
Although a greater proportion of COPD patients who have a low eGFR compared to controls has been reported, the utility of a creatine-based eGFR as a measure of renal impairment in older populations is debateable, particularly in conditions associated with altered body composition. The large proportion of subjects in both groups with impaired eGFR in this study is striking but likely due to the older nature of subjects. The advent of the biomarker era permits the ability to untangle potential anatomical sites for nephropathy. The similar concentration-corrected NGAL and KIM-1 proximal tubular biomarkers lend support to the increased urinary albumin being a primary glomerular injury. We understand this is the first to study these in COPD patients in such a comprehensive manner.
A number of studies have explored different formulae for cystatin C derived eGFR in a variety of populations. The most widely used in an unselected population is the Grubb equation which has been validated against the gold standard (invasive) method of plasma iohexol clearance. Given none of the various equations have compared cystatin C to a gold standard in COPD we opted to present the cystatin C alone. Cystatin C measures glomerular function and it is therefore not all-together unexpected that a marked difference in glomerular damage markers was seen despite similar cystatin C.
The association of glomerular damage to microvas-culopathy has been demonstrated in subjects with other conditions such as hypertension and diabetes, both conditions associated with increased aortic stiffness. In order to interrogate whether aortic stiffness leads to microvascular damage via altered renal haemodynamics, renal doppler sonography permits calculation of the renal artery vascular resistance. Such studies have demonstrated associations of a resistive index to proteinuria in patients with chronic kidney disease with and without diabetes mellitus; to albuminuria and also a measure of aortic stiffness (brachial-ankle) in 150 patients with type 2 diabetes mellitus; and in patients with hypertension, a modest increase in renal resistive index was associated with a greater adjusted relative risk of albuminuria. Fesler and colleagues highlighted the importance of PP (as a marker of arterial stiffness) in longitudinal change in renal decline whilst treating hypertension. Conversely, there is the opposing argument in a biological system - Wang reported the utility of low grade albuminuria to determine progression to hypertension in a non-hypertensive and non-diabetic population over nearly 3 years. This study confirms and highlights the issue in COPD which has not been previously demonstrated.
Increased permeability due to impaired filtration of the barrier may be due to direct podocyte injury but other potential reasons for the association may be mediated via systemic inflammation, hypoxaemia, endothelial dysfunction or increased sympathetic activation, all known to be affected in COPD. Resting oxygen saturations were a significant variable of albuminuria in our study. This has been reported previously in patients with COPD and in other conditions and states such as with altitude in healthy volunteers. Sympathetic activation is known to increase arterial stiffness and raises an interesting consideration that therapeutic modulation, such as with ACE inhibitors, that are CV protective, decrease sympathetic nervous system (SNS) activation and confer renal benefit might have a role in COPD. In a parallel manner, beta blockers inhibit SNS activity and have been shown to improve mortality in COPD although their effect on renal function is as yet unknown. Although there was no association with circulating CRP in this study, systemic inflammation cannot be excluded as a mediator. Similarly, endothelial dysfunction and albuminuria remains a strong contender. The most likely cause is the inter-play of a combination of several factors above or even foetal programming but beyond the scope of this study.
The relationship between arterial stiffness and atheroma is still not well understood and at present conventional risk factors, except BP and age do not account for much of the increased arterial stiffness. However, increased stiffness has many non atherosclerotic deleterious associations including increased pulsatility. Identification of microvascular abnormalities such as increased UACR is important and provides an opportune and influential period to intervene prior to development of macrovascular disease.
Points for Consideration
When considering UACR, it is important to reflect on the spot test which is now accepted over twenty-four hour urine collection and has been validated. UACR was measured contemporaneously from fresh urine. The timing of urinary spot samples is often debated despite low diurnal variation–here morning samples where possible, avoiding first void were collected.
Pack years were not matched however all had a pack year history of greater than 10. All were of European ancestry as this could have affected haemodynamic and glomerular results. Unlike previous studies in patients with and without COPD, we did not demonstrate a difference in AIx as a secondary variable. The Stroke Volume index is derived and the method requires further evaluation.