Alharf, Adel Abdullah
Prevalence of and clinical characteristics associated with microalbuminuria in hypertension.
PhD thesis, University of Glasgow.
Full text available as:
Cardiovascular disease is the leading cause of mortality. As blood pressure is one of the most important risk factors for cardiovascular disease, effective management of hypertension is critical in reducing this risk. In addition to high blood pressure, however, several factors have been identified as predictors of future cardiovascular events. These include high cholesterol, cigarette smoking, obesity and diabetes. Taken together, these traditional risk factors do not entirely explain the risk. Thus, many novel risk factors have been proposed for risk prediction of cardiovascular disease. Microalbuminuria is one such factor.
Microalbuminuria is defined as excretion of albumin in the urine above the normal level but less than gross proteinuria. As excretion of albumin exhibits high variability due to many confounders (such as urinary tract infection and strenuous exercise), diagnosis of microalbuminuria should be ideally based on screening of multiple samples using either 24-hour urine collection or first-morning voids. Much evidence suggests that microalbuminuria is a reflection of generalised endothelial dysfunction. This is supported by the observation that microalbuminuria is strongly associated with cardiovascular disease.
My main aim was to study microalbuminuria in people with hypertension attending specialist clinics. Microalbuminuria has been investigated extensively in diabetes and in patients with renal disease. However, the available information on the association of microalbuminuria with hypertension has many limitations since many studies had small sample size, restricted population or were confounded by potential misdiagnosis of microalbuminuria by the use of single samples. This has led to uncertainty about the prevalence of microalbuminuria in hypertension, where reported prevalence ranges from 4.7% to 58%, and probable underestimation of its clinical significance.
I addressed these issues by conducting a series of studies in 1059 hypertensive subjects attending the Glasgow Blood Pressure Clinic or the Aberdeen Hypertension Clinic. Each patient was invited to provide an early morning urine specimen for the assessment of albuminuria. Urinary tract infection was tested using urine strips and, where positive, samples were discarded. If the first sample showed increased albumin excretion, two further samples were requested. Albuminuria (microalbuminuria or gross proteinuria) was diagnosed when two out of the three samples showed increased albuminuria. Two definitions of microalbuminuria were used in the analysis, a conventional definition with the threshold used by most therapeutic guidelines and a new definition that accounts for low excretion of albumin. All patient information was obtained from case-records.
In the first study, I showed that microalbuminuria by the conventional definition was present in 9.5% of non-diabetic hypertensive subjects without renal impairment. Another 10% of this cohort had microalbuminuria by the new definition. Compared with people with normal urinary albumin, individuals with microalbuminuria by both definitions (n= 786, after excluding those with diabetes or severe renal impairment) had significantly higher blood pressure, higher pulse pressure, increased levels of inflammatory markers, poorer renal function, higher triglycerides levels and used more cardiovascular drugs.
In a second study, the association of microalbuminuria with clinical characteristics was investigated. Subjects with microalbuminuria had increased prevalence of risk factors / co-morbidities such as left ventricular hypertrophy (19.2% in normoalbuminuria versus 29.7% and 34.8% for microalbuminuria by the new and the conventional definitions, respectively), ECG abnormalities and cardiovascular disease. In addition, people with microalbuminuria had higher risk scores for subsequent cardiovascular events using two risk calculators, the Framingham and the Joint British Societies equations. In a subcohort with controlled blood pressure and without co-morbidities or risk, microalbuminuria (by combining the two definitions) was found in 14%. Compared with those with normoalbuminuria, subjects with microalbuminuria had higher blood pressure, poorer renal function, higher blood glucose and higher levels of inflammatory markers although the limited sample size precluded statistical significance.
In a further study, the independent association of microalbuminuria with different risk factors was evaluated using multivariate testing. Systolic blood pressure, serum creatinine, left ventricular hypertrophy and fasting triglycerides were among factors linked with microalbuminuria. The risk of microalbuminuria increased in people with poorly controlled blood pressure. I also found that microalbuminuria was associated strongly with left ventricular hypertrophy [odds ratio 1.87 (95% CI, 1.12 - 3.12) for a composite of both definitions- the combined definition] and cardiovascular abnormalities [odds ratio 1.72 (95% CI, 1.05 - 2.80) for the combined definition].
In a fourth study, the reproducibility of microalbuminuria screening was investigated. I discovered that a large proportion of people who had increased urinary albumin excretion on first sample was categorised as normoalbuminuria based on the result of multiple samples (48% at the Glasgow Clinic and 41% at the Aberdeen Clinic). This indicates that even after controlling microalbuminuria confounders, multiple testing can be recommended for more accurate diagnosis.
In the final study, I demonstrated that subjects with microalbuminuria by both definitions attending the Glasgow Blood Pressure Clinic had relatively high blood pressure and pulse pressure at first visit and subsequently. This finding indicates that subjects with microalbuminuria require particularly rigorous blood pressure management to achieve blood target blood pressure. Furthermore, individuals with microalbuminuria may be at risk for cardiovascular disease greater than that in those with normoalbuminuria since the eventual blood pressure remained higher in these subjects. Together with the observations that microalbuminuria is associated with clustering of cardiovascular risk factors, my finding support the importance of even small increase in urine albumin excretion as an indicator of eventual cardiovascular disease.
In conclusion, microalbuminuria is found in one-fifth of subjects with essential hypertension. Although my investigation was observational, the large sample size, the use of multiple samples and allowance for the effects of potential confounders enhances the precision of the results. Moreover, subjects involved in this study represent a real hypertension population with few restrictions. My findings support the value of microalbuminuria as a tool to identify subjects at high risk for cardiovascular disease. Before routine screening can be recommended, these observations require confirmation in clinical trials and prospective studies with long-term follow up. Linkage of the records of the patients who participated in this series of studies with national morbidity and mortality statistics offers one approach with the potential to test the clinical relevance of my findings.
Actions (login required)