Strengths and limitations of the study
To the authors' knowledge, this is the first study that evaluates the prevalence and severity of CKD in primary care patients with diabetes or hypertension with the use of both serum creatinine assessments (to estimate GFR) and urine albumin excretion (to determine albuminuria). Screening uptake rates are high, providing valid prevalence estimates. The prevalence of stage 3 is greater than that of stages 1 and 2. This is due to the definition of stage 1 and 2 requiring evidence of kidney damage (albuminuria), which is not required for stage 3.
Some methodological issues need particular attention. First, the study is limited by its cross-sectional design. There is no information available about the course of CKD, and factors that may predict rapid worsening of albuminuria or kidney function and subsequent cardiovascular events. Moreover, due to the relatively small samples in this study, associations between different severity levels (of both diabetes and hypertension) and CKD have not been studied. Follow-up data are required on how the identified associations with albuminuria (diabetes, age, and high current blood pressure) and decreased eGFR (age and sex) predict cardiovascular and renal outcome. Second, the study was performed in a limited number of primary healthcare centres, which are possibly not representative of all Dutch family physicians. The practices have affiliations with the Leiden University Medical Centre, which reflects increased awareness of GPs and staff of obtaining valid data and high screening uptake rates. However, this study shows that adherence and commitment to management guidelines with regard to diabetes, hypertension, and screening for CKD can lead to remarkably high screening uptake rates and a high proportion treated with ACE inhibitors/angiotensin receptor blockers in family practice.
Third, the study used the urine albumin-to-creatinine ratio as a measure of kidney damage. However, other measures are available such as urine sediment abnormalities and abnormal findings on imaging studies. Therefore the prevalence estimate of CKD stage 1–2 might be an underestimation, because some cases were missed.
Comparison with existing literature
This study evaluates the prevalence and severity of CKD in primary care patients with diabetes or hypertension with the use of assessments of serum creatinine as well as urine albumin. Some studies reported the prevalence of CKD stage 3–5, but not stage 1–2 (albuminuria), in high-risk primary care patients.13,14 Others assessed the prevalence of CKD in the general population was about 5–7%.7,8 However, the validity of these prevalence estimates was impaired by incomplete eGFR and albuminuria data.7,8 In this study, the focus was on high-risk primary care patients (that is, with diabetes or hypertension) in order to identify the proportion of people with CKD in candidate risk groups.1,15 In patients with hypertension only, the screening uptake rate was lower (87%) than in patients with diabetes (97%). The patients with hypertension who did not attend screening had more cardiovascular comorbidity and used fewer ACE inhibitors and angiotensin receptor blockers. Therefore, the study finding of 21% CKD in patients with hypertension only might be an underestimation of the actual prevalence, since CKD is likely to be highly prevalent in the non-responders with cardiovascular comorbidity and possible undertreatment.
The definition and classification of CKD is made in relation to kidney damage (albuminuria) and decreased kidney function (GFR estimated by the MDRD formula). The distinction between kidney damage and decreased kidney function is important, because prevalence and risk management are different. The overall prevalence of albuminuria in the present study population was about 7%. Albuminuria was much more prevalent in patients with diabetes (14%) than in patients with hypertension only (4%). Age and high systolic blood pressure were also associated with albuminuria. Since albuminuria predicts cardiovascular events, albuminuria is a guide to treatment with ACE inhibitors or angiotensin receptor blockers.4,16
There appears to be some progress in the use of ACE inhibitors/angiotensin receptor blockers in primary care diabetes patients: in the present study sample three-quarter of patients with diabetes and albuminuria were treated with ACE inhibitors/angiotensin receptor blockers compared with about one-third found in previous research.17 However, there is still room for improvement, not only regarding treatment of albuminuria with ACE inhibitors/angiotensin receptor blockers, but also regarding treatment of patients with diabetes and hypertension with antihypertensive medication to prevent CKD and cardiovascular disease.17
The prevalence of decreased kidney function (CKD stage 3–5) was 18%. In contrast to the association with albuminuria, diabetes and high systolic blood pressure were not associated with a decreased estimated GFR. However, statistically significant associations with decreased kidney function were found for age and sex. It is well recognised that kidney function declines with age, but decreased eGFR in older patients does not always reflect kidney disease and increased cardiovascular risk.18,19
Recent data show that cardiovascular morbidity and mortality in an older population aged ≥70 years was only increased in eGFR levels below 50 ml/min per 1.73 m2, but not in levels between 50 and 60 ml/min per 1.73 m.2,20 Cardiovascular management in older patients with decreased eGFR may, therefore, differ from cardiovascular management in younger patients with impaired renal function.21
Although the beneficial role of ACE inhibitors/angiotensin receptor blockers in non-proteinuric kidney disease is less well established,4 there are several reasons that justify identification, close follow-up, and intensified cardiovascular risk management of high-risk patients with decreased kidney function. First, decreased GFR is independently associated with the risk of death and cardiovascular events in young and middle-aged individuals.2,3,22
Second, several adverse pathophysiological consequences such as anaemia, hyperphosphataemia, vitamin D deficiency, and hyperparathyroidism may need treatment or referral. Third, identification of decreased kidney function allows better dosing of drugs excreted by the kidney and avoidance of nephrotoxic drugs such as non-steroidal anti-inflammatory drugs.