NDT Advance Access published online on April 5, 2008
Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfn069
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The detection of advanced chronic kidney disease by surveillance of elevated plasma creatinines—a five-year experience
Department of Renal Medicine, Crosshouse Hospital, Kilmarnock, UK
Correspondence and offprint requests to: Andrew Innes, Department of Renal Medicine, Crosshouse Hospital, Kilmarnock, UK. Tel: +44-1563577358; Fax: +44-1563577987; Email: ainnes{at}aaaht.scot.nhs.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsConclusionsReferences |
|---|
Background. The late referral of patients with advanced chronic renal failure to a nephrologist is multifactorial but also compromises the preparations for dialysis and is prejudicial to their survival on dialysis. Measures that prompt or hasten referral, will allow preparation for dialysis, control of complications and treatment of comorbid conditions.
Methods. In June 2000, a programme was initiated to provide surveillance of plasma creatinines >300 µmol/L on all laboratory requests from general practitioners (GPs) and hospital clinicians in Ayrshire and Arran Health Board in southwest Scotland. Patients already known to the nephrologists were excluded. Results were regularly reviewed and further excluded if the creatinine fell or the patient died. For the remainder, a standard letter was sent to the requesting clinician suggesting renal referral if appropriate. This was to act as a prompt to the general practitioner or hospital clinician. For those referred over the 5-year period, the outcome was analysed in January 2007.
Results. In the first 5 years (June 2000–June 2005) letters were sent regarding 246 patients (median age 76). Fifty-three patients still had reversible ARF or died within 3 months of the letter; seven were already referred. The requesting clinician felt that referral was not appropriate in 56; 23 were being reviewed elsewhere. The programme has led to the referral of 50 patients to the renal service (and 3 to others) but in 54 cases no reply was received and the letter ignored. Of the 50 referred, 17 entered the dialysis programme, 13 of whom had definitive dialysis access (fistula or Tenckhoff catheter) at the start. After a period of outpatient review they have undergone a median of 21 months of dialysis.
Conclusions. Over the 5-year period this programme has detected a cohort of patients who, in general, benefited from nephrological follow-up and dialysis. It may also act as a prompt to clinicians to refer more ‘marginal’ patients and thereby hasten future referral.
Keywords: five; nephrology; programme; surveillance; years
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsConclusionsReferences |
|---|
Late referral of patients with advanced renal disease is associated with significant morbidity and mortality and compromises preparations for dialysis [1–3]. It also incurs greater costs than more timely referral [4,5]. Our own local experience suggests that patients first reach a nephrologist at a relatively advanced stage of their renal disease. The median creatinine (at the time of first referral to a nephrologist) for patients who subsequently started dialysis at Crosshouse Hospital in 1999 was 390 µmol/L (median MDRD eGFR = 14 mL/min/1.73 m2). Measures to enhance and hasten referral will allow preparation for dialysis, control of complications and treatment of comorbid conditions. Programmes to screen for such patients have been suggested previously, mainly looking at hypertensive and diabetic populations [6,7].
In June 2000, a programme was instituted to detect plasma creatinines >300 µmol/L on all laboratory requests from general practitioners (GPs) and hospital doctors in Ayrshire and Arran Health Board in southwest Scotland in patients who did not appear to be known to the local nephrologists. It was to act as an aide-memoire to those requesting clinicians. This article details the findings over the first 5-year period of this detection programme and the status of patients referred by the programme followed up to January 2007.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsConclusionsReferences |
|---|
A surveillance programme was devised to detect plasma creatinines >300 µmol/L on all biochemistry laboratory requests sent by GPs and hospital doctors in Ayrshire and Arran Health Board (population 375 000). The Health Board area is mainly semi-rural with several large towns. It is served by two main hospitals one of which contains the renal unit and inpatient and outpatient renal facilities. The other hospital has outpatient renal facilities and ward consults only, provided by visiting nephrologists from the first hospital. Fifty-eight percent of the total population base of the Health Board is served by the hospital with on-site nephrology; forty-two percent is served by the other hospital that provides all urology services. The surveillance programme commenced in June 2000 and an explanatory letter was sent to all GPs and hospital consultants. Several replies complimenting this initiative were received and no GPs or hospital doctors declined to take part. Participation by clinicians was agreed but with the caveat that only a single letter regarding an elevated creatinine (see below) would be sent with no further enquiry made. All blood samples in the Health Board area are analysed by a single laboratory and the details (name, address, date of birth and requesting clinician) for each sample with a creatinine >300 µmol/L were sent to the author. The start of the programme antedated the K/DOQI classification of chronic kidney disease (CKD) [8] and the use of eGFR but MDRD eGFR results are also given in the results section. Patients already known to the nephrologists were excluded. In a sample month (August 2000) 354 results (with creatinines >300 µmol/L) were received from 90 individual patients. As this is an ongoing programme, a further month was sampled (August 2005) yielding 402 results from 118 patients. Over the same period general laboratory requests for U&Es have risen by 47%.
As the aim was to detect chronic, not acute, renal failure (CRF) the patient details were analysed in batches on average 1 month later. Patients for whom the elevated creatinine was a terminal event, who had died in the interim, were excluded. Similarly, patients with ARF that had reversed in this period and patients who were known to the nephrologists but had samples sent by others were also excluded. For the remaining patients, a single letter was sent to the clinician who had sent the sample. This letter documented that the patient had a creatinine >300 µmol/L but was not known to the local nephrologists, and indeed, I was entirely unaware of the clinical circumstances (other than age and gender) of the patient as well. It suggested that referral or discussion was made regarding nephrological follow-up. Also, if the clinician felt that nephrological input was not appropriate, it was emphasized that the nephrologists would value knowing these particular circumstances too. The liberal nature of our policy towards dialysis was emphasized. As this was acting as an aide-memoire to clinicians and not obligatory, it was agreed that if there was no reply, no further communication was made to the clinician who sent the sample.
Patient demographics, first creatinine >300 µmol/L (and eGFR) and details of the requesting clinicians and their replies (or not), were recorded.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsConclusionsReferences |
|---|
In the first 5 years of this programme (June 2000–June 2005), letters were sent to requesting clinicians on 246 patients (median age 76 years) (Figure 1). Fifty patients were referred to the local nephrology service in response to the letter. In an additional three cases, in response to the letter, the patients were referred not to a nephrologist but to a cardiologist or geriatrician instead. Fifty-three others still had what subsequently turned out to be reversible ARF (n = 35) or died within 3 months of the letter being sent (n = 18). In 56 patients the requesting clinician deemed nephrological referral to be inappropriate usually due to extreme frailty, advanced malignancy, end-stage cardiac failure or dementia. In 54 cases, the requesting clinician failed to reply. The evolution of these patients and of the others in response to letters is shown in Figure 1.
|
Of the 246 patients for whom a letter was sent, 149 (61%) were male and 97 (39%) female. This is similar to the observed sex ratios noted in the incidence of CRF and in dialysis populations both locally and in the UK [9]. In the ‘referred’ patients and ‘no replies’ the percentage of males was higher (70% and 72%, respectively). In the ‘not appropriate’ group, the numbers of males and females were equal. Data of age, initial creatinines (>300 µmol/L) and eGFR for ‘all’ patients, ‘referred’ patients, ‘not appropriate’ patients and ‘no replies’ are shown in Table 1. Statistical analysis was done by the Mann–Whitney U-test. The ‘referred’ patients were significantly younger than ‘not appropriate’ patients and ‘no replies’ (both P < 0.05). The ages of the ‘not appropriate’ group did not differ significantly from those of the ‘no replies’ (P = 0.13). The initial creatinine was similar in all the groups. The initial eGFR was significantly higher in the ‘referred’ patients than in the ‘not appropriate’ patients (P = 0.02).
|
The presenting creatinines (eGFRs), for all patients in our Health Board area (at first presentation to a nephrologist) who went on to require dialysis, for the 4 years before the start of this surveillance programme (1997–2000) were compared with the 4 years immediately afterwards (2001–2004); there are no statistically significant differences. The take-on rate for our Health Board area did not change between 1999 and 2005 (both 112 ppm), whereas the rate for Scotland as a whole rose by 11% (110–122 ppm) in the same period [9].
The details of the specialty of the requesting clinician (GP, medical specialties, surgical specialties) and location (general practice served by or specialty in hospital with or without inpatient nephrology) are shown in Table 2. Most of the patients located by this programme were detected in hospital practice (54% of the total) and the same percentage (54%) was found in the area without the on-site renal unit. Once informed of the patient, the clinicians in the part of the Health Board without on-site nephrology were better at referring the patient. A failure to reply was similar to the population distribution in both areas of the Health Board.
|
The outcomes of the ‘referred’ patients in this 5-year time period were analysed at the start of 2007 (Figure 1). Of the 50 referred patients, 17 had entered the dialysis programme; 4 of whom were treated initially with CAPD/APD. Nine of the thirteen undergoing haemodialysis had a fistula created prior to dialysis. The median duration of clinic follow-up prior to dialysis for those 17 patients was 20 months (range 2–47 months). They dialysed to date for a median of 21 months (range 1–62 months). Six of the seventeen had died by the start of 2007 and those who died dialysed for a median of 15 months (range 1–22 months). Using the risk stratification described by Wright [10,11] based on age and comorbid conditions, 11 of the dialysed patients were high risk and 5 medium risk. Only one patient fell into the low-risk category.
The remainder were followed up in nephrology clinics apart from two who died before they were able to attend. By January 2007, 20 of the 31 had died during clinic follow-up; those dying had a median follow-up of 10.5 months. During clinic follow-up, in addition to the 17 who reached dialysis, 13 others showed progressive renal dysfunction but 18 had either no evidence of progression or improvement in their renal function.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsConclusionsReferences |
|---|
This programme was designed to address the problem of late referral or non-referral of patients with advanced renal failure who are not known to nephrologists [1–5]. While in many cases referral may not be appropriate due to frailty (both physical and mental) and the terminal stages of many illnesses, notable malignancy and cardiac failure, in other cases it is. This initiative was well received by the local clinicians with no objections made and no one declining to take part; indeed several letters of support were sent. This acceptance may have been because, after the initial letter, no further attempt was made to badger or cajole the clinician to reply or refer the patient. This move to maximize acceptance did, however, mean that I was unable to obtain any further details of the patients for whom no reply was received.
The patients with high creatinines, not known at that point to the nephrologists, are predominantly elderly (median age 76 years; the median age of our patients commencing dialysis in 2006 was 64 years). As might be expected, the patients within this group who were then referred were younger. It is impossible to know if these patients, who were referred after receiving the prompting letter, would have been referred, at a later point in time, or even referred at all, had the letter not been sent. Furthermore, it is difficult to know what would have motivate a physician to refer at a later point in time, when even being informed that a patient had a creatinine of >300 µmol/L would not prompt a referral. The oldest patients in the group were deemed not appropriate—the frail elderly, those with terminal malignancy and cardiac failure. Patients in the group ‘no reply’ were similar in age to those considered ‘not appropriate’ suggesting that the clinician failed to reply because he may also have considered the patient ‘not appropriate’ and nephrological input/dialysis futile. This is, however, supposition as this could not be clarified by follow-up. In general, patients detected by this programme are likely to be those who may derive marginal benefits from dialysis (‘marginal’ patients). They are a high-risk population with significant morbidity and mortality not entirely dissimilar to those identified undergoing nursing home haemodialysis [12].
The fact that the majority of the detected patients came from the area without the on-site renal unit suggests that there may be an initial reluctance to refer patients to nephrology services from areas without an on-site nephrological service. It is perhaps not entirely clear to non-nephrologists that attitudes to dialysis are more liberal than they were previously and that conservative approaches to advanced renal failure have changed in recent years. The fact that there was, in many cases, a clear desire to refer in response to a letter indicating the creatinine result would appear to back this up. The considerable number of ‘no replies’ would indicate that there is still some way to go.
Unfortunately this programme did not appear to influence earlier referral of the entire cohort each year who reached dialysis. It also did not change the overall take-on rate, although changes to that are probably multifactorial. It may influence referral of borderline patients but those referred made up a small number of the total starting dialysis (138 between 2001 and 2004) and also the effect may take longer to filter through.
If all the referred patients had reached dialysis in an orderly fashion over the 5 years of the programme, it would have increased our take-on rate by 26.6/million population but clearly this does not represent the true evolution of advanced renal failure and in the follow-up period only 17 did undergo dialysis. The fact that 76% of these had definitive dialysis access when first dialysed suggests that follow-up generated by this programme allowed sufficient time for dialysis preparation and avoided emergency dialysis. Such high- and medium-risk patients have been maintained on dialysis for reasonable periods and can therefore be satisfactorily maintained on dialysis. They are, however, ‘marginal’ patients who, although they did attend clinic for a decent length of time, sustained 20 deaths. Not all patients with advanced renal failure showed further progressive decline. Other studies have shown that only 4% of patients with CKD stage 3 will need dialysis over a 10-year period [13] and in CKD stage 4 only 25% would require dialysis in 5 years [14]. For many of these patients (and all of the studied patients fell into the CKD stage 4 and 5 categories) there is not a relentless decline towards dialysis and the more likely outcome is death, most frequently from vascular disease [13,15,16].
This surveillance programme was started before a general acceptance and introduction of the MDRD eGFR but the use of a creatinine of 300 µmol/L to determine advanced renal failure places all the patients detected in stage 4 and 5 CKD [8]. Furthermore, this programme also antedated the quality and outcome framework in primary care in the UK in 2006 that established registers of patients with CKD stages 3–5 [17]. There is a concern that this has produced an increase in referrals of patients with CKD in the UK leading to pressure on specialist nephrological services and increased patient anxiety without clear proof of benefit [18]. On the other hand, it should also lead to the referral of patients with advanced renal failure who have been detected by this surveillance programme and who, for a variety of reasons, were not referred in the past. Evidence also suggests that the widespread use of eGFR has decreased the number of late referrals [19].
There were a number of problems. The large number of results (over 350) and patients (over 80) each month with markedly elevated creatinines reflects the large and increasing load placed on biochemistry departments. One advantage was that the Health Board area is served by one biochemistry laboratory. The results (and patient details) had to be checked on three cumbersome non-interactive computing systems. For many patients a creatinine >300 µmol/L was a terminal event and therefore, frequently a predictor of impending death. The programme is labour intensive but sustainable though the failure to reply rate of 22% is disappointing.
Over a 5-year period, this surveillance programme has produced a cohort of patients who may have been ‘marginal’ cases but were in general suitable for dialysis. It may act as a prompt to clinicians (who perhaps do not run into a nephrologist on a daily basis) to refer patients they feel may not be dialysis candidates and educate such clinicians regarding future referrals. It may improve or hasten future referral, though the evidence to date would not support that and it may marginally influence rates of nephrological follow-up and dialysis take-on.
|
Acknowledgments |
|---|
I wish to thank Dr D. Boag of the Biochemistry Department, Crosshouse Hospital, for providing the creatinine results, my secretaries for their input, my nephrology colleagues for following up these extra patients and all the local clinicians who agreed to take part.
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsConclusionsReferences |
|---|
- Innes A, Rowe PA, Burden RP, et al. Early deaths on renal replacement therapy: the need for early nephrological referral. Nephrol Dial Transplant (1992) 7:467–471.
[Abstract/Free Full Text] - Roderick P, Jones C, Drey N, et al. Late referral for end stage renal disease: a region-wide survey in the south west of England. Nephrol Dial Transplant (2002) 17:1252–1259.
[Abstract/Free Full Text] - Kessler M, Frimat L, Panescu V, et al. Impact of nephrology referral on early and midterm outcomes in ESRD: EPidémiologie de l’ Insufficance REnale chronique terminale en Lorraine (EPIREL): results of a 2-year prospective, community-based study. Am J Kidney Dis (2003) 42:474–485.[CrossRef][Web of Science][Medline]
- Jungers P, Zingraff J, Albouze G, et al. Late referral to maintenance dialysis: detrimental consequences. Nephrol Dial Transplant (1993) 8:1089–1093.
[Abstract/Free Full Text] - McLaughlin K, Manns B, Culleton B, et al. An economic evaluation of early versus late referral of patients with progressive renal insufficiency. Am J Kidney Dis (2001) 38:1122–1128.[Web of Science][Medline]
- Jungers P. Screening for renal insufficiency: is it worthwhile? Is it feasible. Nephrol Dial Transplant (1999) 14:2082–2084.
[Free Full Text] - Kissmeyer L, Kong C, Cohen J, et al. Community nephrology: audit of screening for renal insufficiency in a high risk population. Nephrol Dial Transplant (1999) 14:2150–2155.
[Abstract/Free Full Text] - National Kidney Foundation K/DOQI. Clinical practice guidelines for chronic kidney disease: evaluation, classification and stratification. Am J Kidney Dis (2002) 39(Supp 1):S1–S266.[CrossRef][Web of Science][Medline]
- Ansell D, Feest TG, Tomson C, et al. UK Renal Registry Report (2006) Bristol, UK: UK Renal Registry.
- Wright LF. Survival in patients with end stage renal disease. Am J Kidney Dis (1991) 17:25–28.[Web of Science][Medline]
- Khan IH, Catto GRD, Edward N, et al. Influence of co-existing disease on survival on renal replacement therapy. Lancet (1993) 341:415–418.[CrossRef][Web of Science][Medline]
- Reddy NC, Korbets SM, Wozniak JA, et al. Staff-assisted nursing home haemodialysis: patient characteristics and outcomes. Nephrol Dial Transplant (2007) 22:1399–1406.
[Abstract/Free Full Text] - Eriksen BO, Ingebretsen OC. The progression of chronic kidney disease: a 10-year population-based study of the effects of gender and age. Kid Int (2006) 69:375–382.[CrossRef][Web of Science][Medline]
- Conway B, Whitworth C, Harty J. Clinical outcomes of patients with stage 4 chronic kidney disease (CKD). SMJ (2006) 51:47.
- Hemmelgarn BR, Zhang J, Manns BJ, et al. Progression of kidney dysfunction in the community-dwelling elderly. Kid Int (2006) 69:2155–2161.[CrossRef][Web of Science][Medline]
- Locatelli F, Pozzoni P. Chronic kidney disease in the elderly: is it really a premise for overwhelming renal failure. Kid Int (2006) 69:2118–2119.[CrossRef][Web of Science][Medline]
- NHS Employers. Contract changes for 2006/07. 2006 www.nhsemployers.org/primary/primary-2450.cfm.
- Giles PD, Fitzmaurice DA. Formula estimation of glomerular filtration rate: have we gone wrong. BMJ (2007) 334:1198–1200.
[Free Full Text] - Bebb CE, Burden RP. Reporting eGFR has benefits. BMJ (2007) 334:1287.
[Free Full Text]
Accepted in revised form: 28. 1.08
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. J Lee and K. Forbes The role of specialists in managing the health of populations with chronic illness: the example of chronic kidney disease BMJ, July 8, 2009; 339(jul08_3): b2395 - b2395. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||