The impact of introducing automated eGFR reporting on uncovering new cases of chronic kidney disease in a University Hospital in Saudi Arabia

Abstract

Objective: To determine the impact of introducing eGFR automated reporting on uncovering new cases of Chronic Kidney Disease. Methods: All serum creatinine (SCr) in adult patients attending outpatient clinics over a two-month period were recorded and eGFR estimated. Cases with a SCr within normal limits but were in CKD stage 3 (<60 mL/min) or higher were recorded and their numbers, percentages and mean ages calculated. Stages 1 and 2 were excluded from analysis because urinary albumin and other urinary abnormalities were not checked. Results: A total of 26,422 SCr from different patients were included. The mean SCr was 92.5 ± 130.9 μmol/ and the mean eGFR was 99.8 ± 32.4 mL/min/1.73 m². Of all the10,601 males with normal SCr, 0.84% were in CKD stage 3 and in all the 14,695 female, 19.24% were in stage 2 in stage. Of all the 14,695 females with serum creatinine in the normal range, 200 (1.36%) were in CKD stage 3. The patients in stages 2 and 3 were significantly older. Conclusion: If our findings are shown to be true for the rest of Saudi Arabia, one could extrapolate that for each 100,000 serum creatinine assayed for males, 840 new cases will be uncovered in CKD stage 3. The corresponding number for females would be 1360 cases.

• eGFR
• CKD
• automated eGFR reporting
• Saudi Arabia

Introduction

Early referral of patients with CKD has been shown to delay the need for dialysis, reduce the number and duration for hospitalization, improve access to renal transplantation and reduce mortality. For new referrals with CKD stages 3–5, the rates of annual eGFR decline slowed from −7.38 before referral to −1.02 mL/min after referral (p < 0.001).1,2 In fact, in patients with diabetic nephropathy, an increase in eGFR was noted compared to those without diabetes (p = 0.034).1

The risk of mortality increases with delay in referral.3 Compared with early referral patients, late referral patients had a 44% higher risk of death at 1 year after initiation of dialysis (HR, 1.44; 95% CI, 1.15–1.80).4 Early referral patients stay on average 7 to 8 days in the hospital as compared to 30 to 31 days for late referral patients.5,6 Moreover, late referral patients are less likely to be put on the waiting list for transplantation (OR, 0.49; 95% CI, 0.41 to 0.59) and are less likely to be given a transplant (HR, 0.65; 95% CI, 0.55 to 0.77).6

Despite this overwhelming evidence of the benefit associated with early referral to a nephrologist, late referrals remain common (in the range of 15–50% of all the cases).7,8 We reported this to be as high as 60% in Saudi Arabia.9

To facilitate and encourage early referral, national and international bodies recommended that renal function is reported by laboratories as eGFR using MDRD formula10 and that the patient is referred when eGFR is <60 mL/min.11

Unrecognized CKD is quite common. Kidney early evaluation program (KEEP), run by the National Kidney Foundation, showed that the prevalence of eGFR of <60 mL/min when screening high-risk individuals was 20.8%12 and in the general population screening (“NHANES III” study) was 4.7%.13 It was similarly reported to be between 4.6% and 5.7% in several European countries.14–16

The patient with early stage CKD is 5 to 10 times more likely to die from a cardiovascular event than progress to ESRD.17

No previous study was done in the Kingdom of Saudi Arabia to impact the introduction of eGFR automated reporting on uncovering new cases of Chronic Kidney Disease (CKD).

Methods

All the serum creatinine assayed in patients attending outpatient clinics over a two-month period (November and December 2011) at King Abdulaziz Medical City, Riyadh biochemistry laboratory, were recorded and eGFR was calculated using the MDRD formula. Patients attending the renal clinics were excluded. In patients who had more than one result during that period of time, the earliest one was considered.

Serum creatinine was assayed with the use of the Jaffe procedure that produced values traceable to the isotope dilution mass spectrometry (IDMS) using the Architect 16000 chemistry analyzer (Abbott, IL).

GFR was estimated using 4-item MDRD [eGFR (mL/min/1.73 m²) = 186 × serum creatinine^−1.154 × Age^−0.203× (0.742 if female)].

All samples with a serum creatinine within normal limits for the local laboratory had their GFR estimated and their CKD status staged according the internationally accepted CKD staging into stages 3 to 5. Stages 1 and 2 were excluded from analysis because urinary albumin and other urinary abnormalities were not checked. The number, gender, percentage of the total and mean age were calculated for each stage. This study has been cleared by our Institution Ethics Review Board for human studies.

Results

A total of 26,422 patients met the criteria for inclusion in the study. These are the specimens assayed for serum creatinine over a two-month period (November and December 2011). The mean age of the whole group was 50.3 ± 16.9 years and the mean serum creatinine was 92.5 ± 130.9 μmol/L and the mean eGFR was 99.8 ± 32.4 mL/min/1.73 m²

There was a total of 10,601 males (mean age 55 years) in the group. After excluding those with serum creatinine levels above the upper limit of the laboratory for adult males (>115 μmol/L), there remained 9280 patients with a mean age of 52.8 ± 16.1 years. Among this group, the mean serum creatinine was 75.7 ± 13.3 μmol/L and the mean eGFR of 104.9 ± 23.5 mL/min/1.73 m².

Of all the male patients with a serum creatinine level within the normal range for the laboratory for adult males, 78 were in CKD stage 3 (0.84%) (mean age of 77 years). No male patient had eGFR that fell within stage CKD 4 or 5 (Table 1).

Table 1. CKD staging of male patients with serum creatinine within the normal range.

Stage	Number of patients	Per cent of total	Mean age (years)
V (<15 mL/min/1.73 m^2)	0	0
IV (15–30 mL/min/1.73 m^2)	0	0
III (30–mL/min/1.73 m^2)	78	0.84	77

Stages 1 and 2 could not be excluded because urinary albumin and other urinary abnormalities were not be checked.

There were 15,818 females (mean age 47.9 years). After excluding those with serum creatinine levels above the upper limit of the laboratory for adult females (>97 μmol/L), there remained 14,695 female patients [mean age of 46.8 (16.4) years]. Among these, the mean serum creatinine was 57.9 (9.2) μmol/L and mean eGFR of 108.4 (23.2) mL/min/1.73 m²

Of all the female patients with a serum creatinine level within the normal range, 200 were in CKD stage 3 (1.36%). (This group had a mean age of 70.3 years.) No female patient had eGFR that fell within stage CKD 4 or 5 (Tables 1 and 2).

Table 2. CKD staging of female patients with serum creatinine within the normal range.

Stage	Number of patients	Per cent of total	Mean age (years)
V (<15 mL/min/1.73 m^2)	0	0
IV (15–30 mL/min/1.73 m^2)	0	0
III (30–mL/min/1.73 m^2)	200	1.36	70.3

Stages 1 and 2 could not be excluded because urinary albumin and other urinary abnormalities were not be checked.

The mean age of the male patients with eGFR <60 mL/min was 26.6 ± 10.4 years older than the male group as a whole (p = 0.001), and the mean age of the female patients with eGFR <60 mL/min/1.73 m² was 22.4 ± 7.9 years older than the female group as a whole (p = 0.001).

Over the two-month period of sample collection, there were a total of 278 patients who had eGFR <60 mL/min/1.73 m² but normal serum creatinine (78 males and 200 females). This amounts to a total of 139 patients per month or 1668 patients per year for King Abdulaziz Medical City in Riyadh.

Discussion

Serum creatinine measurement is not sensitive for detecting mild-to-moderate kidney failure and is significantly affected by age, gender and muscle mass. As such kidney function could fall by as much as 50% or more before the serum creatinine value rises above the upper limit of normal.18 To overcome this, the general recommendation now is that automated eGFR reporting by used.11

The general recommendations are that patients with eGFR <60 mL/min/1.73 m² are best referred to a nephrologist.11 A large number of patients whose serum creatinine levels fall within the laboratory’s normal limits would be discovered to, in fact, have eGFR <60 mL/min/1.73 m² if their renal function is to be reported by the laboratory as eGFR using the 4-item MDRD formula. Thus the implementation this recommendation would result in larger number of patients being referred to nephrologist. One study reported that such an action could result in an increase of 19.1% following the reporting renal function as automated eGFR.19

Another study20 concluded that reporting renal function using eGFR instead of serum creatinine was followed by an overall increase in the monthly referrals rates by 40%. The patients thus referred were significantly more likely to be older and have stage 3 CKD.

In our present study, we also observed that the patients that would be likely to be referred after such a reporting policy will increase by 1.36% in females if all patients with eGFR <60 mL/min is the cutline referral policy. For males, this increase would be 0.84%. The mean age of the male patients with eGFR <60mL/min was 26.6 ± 10.4 years older than the male group as a whole (p = 0.001), and the mean age of the female patients with eGFR <60 mL/min/1.73 m² was 22.4 ± 7.9 years older than the female group as a whole (p = 0.001).

When this reporting policy was introduced in Australia in 2005, a significant increase in people who “did not attend” (DNA) the clinic appointments was noted (except in those with diabetes mellitus).11

In a recent Canadian study, it was found that automated eGFR reporting was associated with a significant rise in the number of referrals and with a 33% rise in patient waiting time with older patient being referred.21

Another study showed a sevenfold increase in referrals after the introduction of automated eGFR reporting system.22 However, this reduced back to a rate similar to that seen in our study, following the introduction of a referral assessment service.20

In conclusion, in our institution, among a total of 26,422 patients included during the two-months period, 278 of the patients whose serum creatinine fell within the normal range proved to have eGFR <60 mL/min, 78 were males and 200 females or 139 new patents every month (1668 patients per year).

If this proves true throughout Saudi Arabia, one could extrapolate that for each 100,000 serum creatinine assayed for males, 840 new cases will be uncovered who are in CKD stage 3. For 100,000 serum creatinine assayed for females, the numbers would be 1360.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.