Q J Med 2001; 94: 261-265
© 2001 Association of Physicians
Renal impairment in elderly patients with hypertension and diabetes
From the Renal Unit, Kings College Hospital, London, UK
Received 5 November 2000 and in revised form 14 March 2001
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
We investigated the prevalence of renal impairment in individuals with known hypertension or diabetes aged 50–75 years in two South London General Practices. We initially interrogated the practice and hospital biochemistry databases for each individual's most recent serum creatinine. Individuals with no result recorded in the previous year were then invited for screening: 189/365 (51.8%) attended. Data were collected on 821 of a total potential population of 997. Taking a serum creatinine of 120 mmol/l as the upper limit of normal, the overall prevalence of renal disease in this population was 8.4%: 6.1% in the hypertensives, 12.6% in the diabetics and 16.9% in those with both. Significant proteinuria (
2+)
was present in 3.9% of the total population: 2.2% of hypertensives,
8.3% of diabetics and 3.9% of those with both. At screening,
44.5% of individuals had inadequately controlled blood pressure. Renal
impairment is common in this population at high risk of renal disease. Screening
for renal disease in this population is simple, safe and gives a high yield
of positive results. |
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Late referral of renal disease to the nephrologist is associated with increased morbidity and mortality, as well as decreased psychological and social well-being of the patient once commenced on renal replacement therapy.1– 4 Much late referral is unavoidable, being associated with either diseases which have an acute onset or the insidious advance of renal disease which, being relatively asymptomatic, does not lead the patient to present for medical care.5 Other late referral is, however, the result of missed opportunities for screening in high-risk groups or overlooked biochemical results.6
Few data exist on the prevalence of early renal disease in the UK. These data are necessary if there is to be informed debate as to the need for formal screening for renal disease. Recent audit data suggest that existing guidelines on screening for renal disease in hypertensive and diabetic populations are frequently not followed and, when they are, action is not always taken.7 This study provides some data as to the likely prevalence of renal impairment among a population of individuals at high risk of renal disease.
The first report of the National Screening Committee set out guidance on the standard of proof necessary to underpin any screening programme.8 The essence of these guidelines is that: (i) the condition should pose an important health problem and its natural history should be understood; (ii) the population in which the disease occurs should be well defined and its attitude to screening known; (iii) the test(s) for the condition should be sensitive, specific and acceptable to the target population; (iv) the effectiveness of early treatment should be proven and it should be both widely available and acceptable; (v) the programme should be evaluated in terms of cost/benefits and take-up rates.
Any medical condition that warrants either a local or a national screening programme needs to fulfil the criteria set out by the National Screening Committee. End-stage renal failure is an important health problem, with an annual cost of approximately £25 000 per patient treated with dialysis as well as a substantial morbidity and mortality. There are numerous causes of renal failure, and the pathophysiology of the progression to ESRF for many of these conditions is well understood. Certain populations are known to be at increased risk of renal disease, including individuals with diabetes and/or hypertension.
We performed a two-phase cross-sectional and screening study to ascertain the prevalence of, and feasibility of screening for, renal impairment among individuals at high risk of renal disease in two General Practices in Southeast London. Our intention was to answer four questions: (i) of the high-risk patients identified how many had been tested for renal disease in the last year? (ii) how feasible was it to screen those who did not already have a test result? (iii) what would be the prevalence, i.e. the yield of positive results, among the screened cohort? (iv) what is the prevalence of renal impairment among this high-risk population?
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Local research ethics committee approval was obtained. The electronic databases of two South London General Practices were reviewed to identify all patients between 50 and 75 years of age known to have hypertension or diabetes. These patients were considered as being at high risk of developing renal disease, while being young enough to be referred and accepted for nephrological care and, where necessary, renal replacement therapy.
Further information elicited from the database included results of the most recent serum creatinine test and urine strip test for proteinuria, as well as the patient's home address.
All patients without a serum creatinine result recorded in the last year were identified. These individuals were then checked against the local hospital biochemistry database by name and date of birth, and the latest creatinine result, if recorded in the last year, was added to the study database. These individuals constituted a cohort of high-risk individuals who had been tested either opportunistically or purposefully by either the GP or a hospital physician.
The second group—patients with no creatinine result recorded—were then invited by letter to attend for screening. Where patients failed to attend, a second letter was sent. If patients failed to respond to a second invitation, no further efforts were made to contact them.
All patients were given the opportunity of telephoning to arrange a home visit. Participants were screened in either their GPs surgery, the hospital or in a few cases their own home. At the screening visit, the nature and purpose of the study were explained, and written consent obtained. Basic demographic data were recorded and bloods taken for biochemistry. Clean-catch urine samples were either brought in by the patient or provided at clinic, and were strip-tested for the presence of protein and glucose using a 60 s reagent stick. Subjects also had their sitting blood pressure recorded using an electronic sphygmomanometer after at least 5 min rest. Data on serum creatinine and proteinuria from the screened cohort were added to those collected previously from the GP and hospital databases to calculate the aggregate prevalence of renal impairment in this population.
After each individual had been given two opportunities to attend for screening, the biochemistry database was re-checked to see if any non-attendees had had blood tests taken during the course of the study. These results were then added to the database to make the prevalence data more complete. The time from initial data collection to study completion was exactly 6 months. Since the data at initial extraction covered a period of 1 year, these data cover creatinine and protein dip-test results undertaken over an 18-month period.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The initial interrogation of the GP databases revealed 997 individuals between the ages of 50 and 75 years with a history of either diabetes, hypertension or both, out of a total population of this age of 5113. Of these, 731 were hypertensive (14.3%), 205 were diabetic (4%) and 61 had both diabetes and hypertension (1.2%). These figures represent only the known hypertensives and diabetics in these practices, where screening for these conditions was, at the time of the study, opportunistic. Results were available from the GP or hospital databases for 394 (53.9%) of the hypertensive individuals, 184 (89.8%) of diabetics and 54 (88.5%) of the individuals with both diseases. Of the 632 available creatinine results, the GPs held 564, and 68 were found on interrogation of the hospital database.
Invitations to be screened were therefore sent out to 365 individuals: 337 hypertensives, 21 diabetics and seven individuals with both diseases.
In total, 189 (51.8%) individuals attended for screening:
177 (52.5%) of the invited hypertensives, seven (33.3%)
of the diabetics and five (71.4%) individuals with both diseases.
In total, data on the serum creatinines of 821 (82.3%) of
the entire cohort were available at study end (Table 1
). Of those not attending for screening, two died
during the study period, 25 actively declined screening and 149 made no contact
with the GP or researchers during the period of the study.
|
The mean age of the entire cohort was 63.6 years, and 53% were female. Table 2
shows the basic demographic
and mean serum creatinine data for the individuals who had existing data and
those who were screened. There were no statistically significant differences
between the two groups in terms of age, gender or creatinine results, even
when these were examined in disease-specific strata. Ethnicity monitoring
within the practices was incomplete, and data on ethnicity of the group are
not presented.
|
Aggregate data on serum creatinines >120 & thinsp;mmol/l taken from the database searches and the screened cohort, are presented in Table 3
, both overall and for each
disease group. We used 120 mmol/l as the local upper limit of normal
for serum creatinine. While this is a fairly crude estimate of renal function
in the absence of age and weight data, it provides a reasonable tool for estimating
the prevalence of renal impairment in a population. The estimates of prevalence
of renal impairment in the hypertensive, diabetic and both disease populations
are thus 6.1%, 12.6% and 16.9%, respectively.
|
Some 62% of hypertensives had a systolic blood pressure of >150 mmHg, and 47.5% had a diastolic blood pressure of >90 mmHg (the British Hypertensive Society recommends a minimum level of control for audit purposes <150/90 mmHg). 9 Blood pressure targets for diabetics are slightly lower (<140/90 mmHg), the systolic target being achieved in two and the diastolic in six of the seven diabetics screened. Of the five individuals with both diabetes and hypertension, all had systolic blood pressures >140 mmHg, and two had diastolic pressures of >90 mmHg (Table 4
).
|
Proteinuria data are presented in Table 5
.
Diptest results for proteinuria were available in 335 individuals, 188 from
the screened cohort. One patient refused to give a specimen. Proteinuria was
most common in those patients with diabetes and least common in those with
hypertension. Proteinuria was present in 7.9% of the hypertensives
with a result, 29.2% of the diabetics and 20% of individuals
with both diseases. Significant proteinuria (>2+ on dip test)
was present in 2.2%, 8.3% and 5.7% of the hypertensive,
diabetic and ‘both’ disease groups, respectively.
|
One interesting observation was the de novo diagnosis of diabetes mellitus in three screened individuals who had previously been thought to have only hypertension. All three were White, and two were female. Only one patient of the cohort found to have renal impairment was known to the renal unit prior to the start of this study.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The four broad objectives of the study were met. Approximately half of the hypertensives, and 90% of the diabetics and individuals with both diseases, had been tested for renal function in the year preceding the study. More than half of those invited to screening attended: all of those agreed to a blood test and almost all (99.5%) had a urinalysis. The yield of positive results (serum creatinine >120 mmol/l) was 7.4% among hypertensives and 14% among diabetics. Together with the pre-existing results, this gives prevalences of renal impairment in these high-risk populations of 6.1% (hypertensives), 12.6% (diabetics) and 16.9% (diabetes and hypertension).
A previous audit of General Practice databases and notes found that 53.1% of patients aged 50–75 with hypertension or diabetes had a creatinine recorded in the preceding 24 months. 7 The prevalence of renal disease was estimated to be 10.6% and 12% in the hypertensive and diabetic populations, respectively. The estimate of disease burden in the present study was different for the hypertensive population at 6.1%, but was similar for the diabetic population at 12.6%. The difference in disease estimates for the hypertensive cohorts was perhaps due to the fact that the present study used prospective screening on otherwise apparently healthy individuals, whereas the previous study audited data on individuals who had bloods taken after visiting their GP.
Urine strip testing plays an important role in disease identification and in quantifying disease severity. On its own the presence of protein in the urine is non-diagnostic since it is not disease-specific, and is not always present in renal disease. In this study, only four (5.8%) of the 69 patients with a raised creatinine exhibited proteinuria, suggesting that proteinuria is not, on its own, a good method of testing for renal disease in this population. The clinician should however be aware that proteinuria, in the absence of hypercreatinaemia, may be associated with early renal impairment in diseases such as diabetes.
Good blood pressure management is important in preventing the progression
of renal disease.10,11 Guidelines on suitable target
blood pressures for individuals with diabetes and hypertension are available
from the British Diabetic Association (now Diabetes UK)12 and the British Hypertensive
Society.9 As shown
in Table 4
, data on blood
pressures taken during screening demonstrate that (supporting data from
other studies8,13) these targets are seldom
achieved. The implementation of clinical governance will require that General
Practices improve their information technology capabilities so that audit
and evaluation of the management of chronic disease can be undertaken regularly.
Early identification of renal disease is important. People with renal impairment are more likely than the general population to have cardiovascular disease, and thus suffer disproportionate morbidity and mortality.14 Early detection and treatment of some renal diseases (especially those associated with diabetes and hypertension) may permit the retardation, or even prevention, of the progression of renal disease.10,11, 15 Therefore early identification of these diseases may have important economic16 (as yet unproven) and quality of life implications.3, 17
Decisions regarding the utility of implementing screening programmes for disease are contingent on a number of premises. The most important of these are: the prevalence of the disease in defined populations; the sensitivity, specificity and acceptability of the screening tests; the morbidity and mortality associated with the disease; the availability and utility of established treatment for the disease; and the economic benefits of treating the disease in its early stages. This study has demonstrated the relatively high prevalence of early renal disease in a high-risk population, and the relative ease and acceptability of screening. The generalizability of this data to the wider UK population remains to be established. Further research is needed to ascertain the economic and health advantages of both early detection of renal disease, as well as the utility of early referral to the nephrologist.
|
Acknowledgments |
|---|
We thank Dr G. Adam, Dr H. Cotton and Dr B. Amir-Ansari. Funded with an educational grant from Roche Pharmaceuticals.
|
Notes |
|---|
Address correspondence to Mr P.A. Ellis, Renal Administration, King's College Hospital (Dulwich), East Dulwich Grove, London SE22 8PT
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Jungers P, Zingraff J, Albouze G, Chauveau B, Page T, Hannedouche, Man NK. Late referral to maintenance dialysis: detrimental consequences. Nephrol Dial Transplantation1993; 8: 1089–193.
2.
Ellis PA, Reddy V, Bari N, Cairns H. Late referral of
end-stage renal failure. Q J Med1998;
91:727–32.
3.
Sesso R, Yoshihiro MM. Time of diagnosis of chronic renal
failure and assessment of quality of life in haemodialysis patients.
Nephrol Dial Transplant1997; 12:
2111–16.
4. Roubicek C, Brunet P, Huiart L, Thirion X, Leonetti F, Dussol B et al.Timing of nephrology referral: influence on mortality and morbidity. Am J Kidney Dis2000; 36:35–41.[Web of Science][Medline]
5.
Innes A, Rowe PA, Burden RP, Morgan AG. Early deaths
on renal replacement therapy: the need for early nephrological referral.
Nephrol Dial Transplant1992; 7:
467–71.
6.
Khan IH, Catto GRD, Edward N, Macleod AM. Chronic renal
failure: factors influencing nephrology referral. Q J Med
1994; 87:559–64.
7.
Kissmeyer L, Kong C, Cohen J, Unwin RJ, Woolfson RG,
Neild GH. Community nephrology: audit of screening for renal insufficiency
in a high risk population. Nephrol Dial Transplant1999; 14:2150–5.
8. National Screening Committee. First Report of the NSC. London, Department of Health, 1998.
9. Ramsay LE, Williams B, Johnston GD, MacGregor GA, Poston, Potter JF, et al. Guidelines for the management of hypertension: report of the third working party of the British Hypertension Society. J Hum Hypertension1999; 13: 569–92.[Web of Science][Medline]
10. Zucchelli P, Zuccala A, Borghi M, Fusaroli M, Sasdelli M, Stallone C, et al. Long-term comparison between captopril and nifedipine in the progression of renal insufficiency. Kidney Int 1992; 42:452 –8.[Web of Science][Medline]
11.
Klahr S, Levey A, Beck GJ, Caggiula AW, Hunsicker L,
Kusek JW, Striker C. The effects of dietary protein restriction and blood
pressure control on the progression of chronic renal disease. N Engl
J Med1994; 330:
877–84.
12. Viberti GC, Marshall S, Beech R, Brown V, Derben P, Higson N, et al. Report on renal disease in diabetes. Diabet Med1996; 13: S6–12.[Web of Science][Medline]
13.
Dasgupta I, Madeley RJ, Pringle MA, Savill J, Burden
RP. Mangement of hypertension in patients developing end-stage renal
failure. Q J Med1999; 92:
519–25.
14.
Levey AS, Eknoyan G. Cardiovascular disease in chronic
renal disease. Nephrol Dial Transplant1999;
14:828–33.
15. Trocha AK, Schmidtke C, Didjurgeit U, Muhlhauser I, Bender R, Berger M, Sawicki PT. Effects of intensified antihypertensive treatment in diabetic nephropathy: mortality and morbidity results of a prospective controlled 10-year study. J Hypertens1999; 17:1497–503.[Web of Science][Medline]
16.
Ismail N, Neyra R, Hakim R. The medical and economical
advantages of early referral of chronic renal failure patients to renal specialists.
Nephrol Dial Transplant1998; 13:
246–50.
17. Klang B, Bjorvell H, Berglund J, Sundstedt C, Clyne N. Predialysis patient education: effects on functioning and well-being in uraemic patients. J Adv Nurs1998; 28:36–44.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E.B. Jude, S.G. Anderson, J.K. Cruickshank, A. Srivatsa, N. Tentolouris, R. Chandrasekaran, R. Gokal, and A.J.M. Boulton Natural history and prognostic factors of diabetic nephropathy in type 2 diabetes QJM, June 1, 2002; 95(6): 371 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ishida, H. Ishida, M. Narita, T. Sairenchi, Y. Saito, H. Fukutomi, H. Takahashi, K. Yamagata, and A. Koyama Factors affecting renal function in 119 985 adults over three years QJM, October 1, 2001; 94(10): 541 - 550. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||