Comparison of 12-Hour Creatinine Clearance and Estimated Glomerular Filtration Rate in Renal Transplant Recipients

Abstract

Glomerular filtration rate (GFR) is an essential clinical assessment of renal function post-renal transplantation. Creatinine clearance (CrCl) measured over 12 h and estimated GFR (e-GFR) (calculated by the Modification of Diet in Renal Disease equation) were compared in 28 stable renal transplant recipients (RTRs). This single center study included 14 African American (AA) and 14 Caucasian (CC) recipients. The 12-h creatinine clearance (CrCl-12 h) was determined by monitored urine collection and by e-GFR on two occasions (two phases) separated by at least 2 weeks. Statistics included mixed model analysis of CrCl-12 h and e-GFR relative to race, phase, and difference between parameters. In the first phase, the e-GFR was higher in AA males (58.4 ± 14.8 mL/min) than the CC males (46.2 ± 10.2 mL/min) (p = 0.032), whereas the CrCl-12 h of AA males (70.8 ± 8.7 mL/min) and CC males (63.3 ± 21.7 mL/min) was not different (p = 0.740). During the second phase, the e-GFR in AA and CC RTRs was 55.4 ± 10.1 mL/min and 47.6 ± 10.7 mL/min (p = 0.117), respectively, whereas CrCl-12 h in AAs was 64.71 ± 17.9 mL/min and in CCs was 62.0 ± 14.9 mL/min (p = 1.000). The CrCl-12 h was higher than the e-GFR (p < 0.001) irrespective of race or phase. CrCl-12 h was not different on both occasions (p = 0.289) in all the patients. CrCl-12 h was consistently greater than e-GFR. The difference between these e-GFR estimates may have an importance in the care of RTRs.

• creatinine clearance
• e-GFR
• race
• renal transplant

INTRODUCTION

Clinical assessment of graft function following a renal transplant is imperative to monitor acute rejection and chronic allograft dysfunction as well as to adjust the doses of renally eliminated medication. Glomerular filtration rate (GFR) is most widely used and considered as the best parameter to estimate clinical renal function. Inulin clearance is the gold standard for measurement of GFR since this substance is freely filtered by the kidney without undergoing metabolism, tubular secretion, or absorption.¹ Other similarly handled markers labeled with radioisotopes also provide reliable GFR estimates. These include 99 Technetium pentetic acids (DTPA) and Iodine 125 labeled iothalamate.² Unfortunately, none of these methods can be easily implemented in routine clinical practice since they are cumbersome, expensive, and time-consuming or involve radioisotopes.²

A number of equations to estimate GFR using creatinine, an endogenous waste product of muscle metabolism which is primarily cleared by glomerular filtration, are employed in clinical practice. These formulae included the Cockcroft–Gault and Jeliffe equations, which add selected patient demographics to the serum creatinine. The more recently devised Modification of Diet in Renal Disease (MDRD) equation has gained widespread acceptance and commonly used in the clinical arena to calculate estimated GFR (e-GFR). Creatinine clearance (CrCl) using timed urine collection typically of 12 or 24 h and the serum creatinine have remained a main stay in the clinical assessment of glomerular filtration.^3,4 CrCl, however, has the potential to overestimate glomerular filtration since tubular secretion of creatinine increases as renal function declines.^5–7 Other limitations with CrCl measurement are improper collection of timed urine specimens, increased secretion of creatinine due to calcineurin-induced nephrotoxicity, or dietary effects on creatinine.^6,8–9 If urine is collected properly CrCl can provide a fairly accurate estimation of GFR.^10–13 The MDRD equation which was developed for patients with chronic kidney disease also has its inherent limitations in estimating GFR.^14,15 The accuracy of MDRD derived e-GFR is best in patients with GFR less than 60 mL/min.^16,17 If the GFR is above this range, underestimation has been reported.^18–20 The MDRD equation also underestimates the actual GFR in elderly patients and provides overestimates in patients with cirrhosis or who are malnourished.²¹ Thus, accurate determination of GFR in the clinical setting remains challenging, yet critically important during the post-renal transplant period.

The objective was to compare the CrCl-12 h under clinical research unit condition with e-GFR determined using the four-factor MDRD equation in Caucasian (CC) and African American (AA) renal transplant recipients (RTRs) with stable graft function. We also evaluated the effect of race on the disparity between CrCl-12 h and e-GFR and assessed the reproducibility of CrCl-12 h in both the races.

MATERIALS AND METHODS

This was a single center, prospective study conducted in the University at Buffalo Kidney and Pancreas Transplant Unit at Erie County Medical Center (ECMC), Buffalo, NY. This study was approved by the Institutional Review Board (IRB) of University at Buffalo and all patients completed an informed consent. The research protocol and informed consent were in adherence with the Declaration of Helsinky. The study included 28 males at least 6 months post-renal transplantation. The patients were clinically stable and receiving steady state regimens of cyclosporine and a mycophenolic acid preparation as the chronic immunosuppressive regimen. This was a substudy of longitudinal pharmacological evaluation of the chronic immunosuppressive actions of cyclosporine and mycophenolic acid that consisted of two phases separated by at least 2 weeks. During the first phase (phase 1), patients were on mycophenolate mofetil. In the second phase (phase 2), patients were on mycophenolic acid sodium. Patients were admitted to the ECMC Renal Research Unit. They voided before starting urine collection that had always begun at 6 AM. A comprehensive metabolic panel including serum creatinine was included at the start of the study. Study patients remained in the Renal Research Unit over the 12-h period and urine was collected for 12 h under close supervision by research personnel. The patients were encouraged to drink bottled water over the study duration and the intake was recorded, and they were fed three meals during the course of the study. Serum creatinine and urine creatinine were analyzed by the ECMC Clinical Laboratory using enzymatic colorimetric assay (Creatinine plus; Roche Diagnostics Indianapolis, IN, USA) with an inter-assay coefficient of variation <10%. The e-GFR was calculated using the four-factor MDRD equation (e-GFR = 175 × Serum creatinine – 1.154 × Age – 0.203 × [1.212 if African American] × [0.742 if female]).¹⁷ The measured CrCl-12 h was calculated as follows: CrCl = Creatinine urine × Volume urine (mL)/Creatinine serum × Time (h) × 60. The inclusion criteria were (1) male RTR ≥6 months; (2) age >21–70 years; (3) stabilized on the same dose of immunosuppressive drugs for at least 7 days prior to study; (4) serum creatinine ≤3.0 PubMed mg/dL (265.2 μmol/L) with no change greater than 0.25 mg/dL (22.15 μmol/L) in prior two clinic visits; and (5) white blood cell count ≥3000/mm³ (3 × 109/L) and hemoglobin ≥8.0 g/dL (80 g/L). The exclusion criteria were (1) significant gastrointestinal disease interfering with drug absorption; (2) infection 2 weeks prior to study; (3) acute rejection episode 12 weeks prior to study; (4) significant cardiovascular, hematologic, hepatic, psychiatric, neurologic, or oncologic disease.

The data were summarized using the mean, median, and standard deviation for continuous end points and using frequencies and relative frequencies for categorical variables. All the data are expressed as mean plus or minus the standard deviation. In order to assess the differences relative to race and phase, a mixed model with phase, race, phase × race interaction, and random patient effect was used for continuous variables. Diagnostic plots were used to assess model assumptions and identify the need for any data transformation. Racial comparisons were made within each phase using least square means with a Bonferroni correction for multiple comparisons. The racial and phase pair-wise comparisons for categorical end points utilized Fisher’s exact test. A mixed statistical model was used to evaluate the association between e-GFR and CrCl-12 using race, study phase, the interaction of race × phase as predictors relative to the outcome of e-GFR. All the analyses were conducted using SAS statistical software (ver. 9.2; SAS, Cary, NC, USA) with a nominal significance level of 0.05.

RESULTS

A total of 28 RTRs, 14 AA and 14 CC, enrolled in this study. One AA could not participate in phase 1. Patient demographics are summarized in Table 1. Demographics did not change between Phase 1 and phase 2. In phase 1, the e-GFR for AA (n = 13) and CC RTRs (n = 14) was 58.4 ± 14.8 mL/min and 46.2 ± 10.2 mL/min (p = 0.032), respectively (Table 2). In this phase, the 12-h CrCls of African Americans and Caucasians were 70.84 ± 8.7 mL/min and 63.3 ± 21.7 mL/min (p = 0.740), respectively. During phase 2, the e-GFR in African American males (n = 14) was 55.4 ± 10.1 mL/min and 47.6 ± 10.7 mL/min in Caucasian males (n = 14) (p = 0.177). During this phase, the 12-h CrCl in African Americans (n =14) was 64.7 ± 17.9mL/min and 62.0 ± 14.9 mL/min in Caucasians (n = 14) (p = 1.000). When the e-GFR was assessed across phases and race, an overall race effect (p = 0.028) was found with the pair-wise comparison attributed to differences between races during phase 1 (p = 0.032). There was no significant effect of phase observed for CrCl (p = 0.289) or e-GFR (p = 0.820). A positive association between e-GFR and CrCl-12 h was found with a significance trend noted (p = 0.066). A mean difference of 13.11 (±2.68) between CrCl-12 h and e-GFR was significant (p < 0.001) with CrCl providing the greater measurement irrespective of race or phase. We also evaluated GFRs estimated by Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation (e-GFR = 141 × min(Scr/k, 1)α × max(Scr/k, 1) – 1.209 × 0.993 Age × 1.018 (if female) × 1.159 (if black), where Scr is serum creatinine in mg/dL, k is 0.7 for females and 0.9 for males, α is −0.329 for females and −0.411 for males, min indicates the minimum of Scr/k or 1, and max indicates the maximum of Scr/k or 1).¹⁴ There was no difference between GFR estimated by MDRD and CKD-EPI equation.

Table 1.  Demographics of renal transplant recipients during treatment phases.

	Phase 1			Phase 2			Between phase
Parameters	AAM	CCM	p-Value	AAM	CCM	p-Value	p-value
Age (years)	51.8 (9.4)	52.4 (10.5)	0.880	51.8 (9.4)	52.4 (10.5)	0.880	NS
Time post transplant (years)	5.2 (3.4)	7.7 (4.5)	0.118	5.2 (3.3)	7.7 (4.6)	0.117	NS (0.160)
Total weight (kg)	97.9 (26.2)	102.6 (16.8)	0.592	97.9 (27.4)	102.4 (16.9)	0.603	NS (0.889)
Lean body weight	64.8 (7.6)	68.8 (5.0)	0.114	64.7 (8.0)	67.8 (4.3)	0.219	NS (0.376)
BMI (kg/m^2)	32.3 (8.2)	33.4 (6.0)	0.676	32.2 (8.4)	33.4 (6.2)	0.675	NS (0.837)
Albumin (g/dL)	4.1 (0.3)	4.3 (0.4)	0.154	4.2 (0.2)	4.2 (0.4)	0.859	NS (0.949)
(g/L) SI	41 (3)	43 (4)		42 (2)	42 (4)
Creatinine (mg/dL)	1.7 (0.4)	1.7 (0.5)	0.912	1.7 (0.3)	1.7(0.4)	0.973	NS (0.795)
(μmol/L) SI	150.3 (35.4)	150.3 (44.2)		150.3 (26.5)	150.3 (35.4)
Hbg (g/dL)	12.5 (1.2)	12.7 (1.4)	0.772	12.7 (1.6)	12.4 (1.6)	0.625	NS (0.943)
(g/L) SI	125 (12)	127 (14)		127 (16)	124 (16)
Cyclosporine (ng/mL)	128.5 (49.6)	125.6 (48.7)	0.887	135.1 (74.8)	110.8 (29.1)	0.242	NS (0.662)
(nmol/L) SI	106.9 (41.3)	104.5 (40.5)		112.4 (62.2)	92.2 (24.2)
Prednisone (% yes)	46.2%	28.6%	0.440	46.2%	28.6%	0.440	NS

Note: AAM, African American male; CCM, Caucasians male; BMI, body mass index; Hbg, hemoglobin; NS, nonsignificant; SI, The International System of Units.

Table 2.  GFR estimated by MDRD equation and CrCl during MMF and MPA phases of the study.

	Phase 1			Phase 2
GFR	AAM	CCM	p-Value	AAM	CCM	p-Value	Between phase p-value
e-GFR (mL/min)	58.4 ± 14.8	46.2 ± 10.3	0.032	55.4 ± 10.1	47.7 ± 10.7	0.177	0.820
Measured CrCl-12 h (mL/min)	70.8 ± 8.8	63.3 ± 21.7	0.740	67.2 ± 15.9	62.0 ± 14.9	1.000	0.289

Notes: p-Values are Bonferroni adjusted. GFR, glomerular filtration rate; e-GFR, estimated GFR using MDRD equation; CrCl-12 h, creatinine clearance over 12 h; MDRD, Modification of Diet in Renal Disease; AAM, African American male; CCM, Caucasian male.

*Overall difference between the e-GFRs of two phases.

DISCUSSION

GFR is a critical clinical measurement of renal function and provides an important index of health and disease progression.^22,23 Many drugs are renally excreted and GFR is used to guide for the adjustment of medication regimens. Knowledge of the renal function remains paramount in clinical pharmacology and patient management. Direct determination of GFR is difficult because it requires infusion of radiolabeled markers or nonlabeled clinically inert compounds that are not commonly measured in clinical laboratories. The restrictions of these techniques have resulted in widespread use of estimates of GFR using serum creatinine, an endogenous waste product of muscle metabolism. The renal excretion of creatinine can vary in different clinical scenarios. In spite of the limitations, estimation of GFR is an essential clinical assessment parameter to monitor patients with kidney disease including RTRs.²⁴

Repeat measurement of CrCl and its relationship to e-GFR in RTRs including racial subpopulations has not been reported. These data provide an assessment of CrCl-12 h and e-GFR in closely monitored clinically stable African American and Caucasian RTRs. CrCls were similar on both occasions, even though urine was uniformly collected during the daytime, when chance of variation in day to day solute intake is high. The consistent access to food and fluid for all patients on both occasions, however, should have obviated this variation. Precisely measured CrCl during a 12-h urine collection was highly reproducible in the stable RTRs of both races. This may be attributable to careful urine collection, essentially identical body mass indices of patients²⁵ during both the phases of study and stable renal function. The consistency of the results suggests that any phasic excretion of creatinine is obviated by this length of urine collection. Further, these subjects were studied in a consistent time period which should have abrogated the less than 20% diurnal variations of GFR.^26,27 The reproducibility of this CrCl measurement in the less than 24-h samples supports the notion that carefully done shorter urine collection can produce reliable result in the clinical setting. General agreement between 12-h and 24-h CrCl is indeed well recognized.^28,29 Hypothetically, some differences might be found if the 12-h urine collection was limited to those times of the day (about 9 am to 9 pm) when GFR is usually at its highest. Since we have only 12-h data, we cannot exclude the possibility that 24 CrCls would have been somewhat lower in this subject population. The circadian variation in creatinine excretion is retained in transplant recipients but it was minimal.³⁰ Among the various methods to estimate GFR, the MDRD equation has provided the most reliable estimation in renal transplant patients.^2,31 There are certain clinical conditions where CrCl should be considered instead of e-GFR such as in patients with substantial muscle wasting, rapidly changing kidney function, and GFR greater than 60 mL/min.¹⁷

The CrCl was significantly different from e-GFR irrespective of race. The measured CrCl is consistently elevated from 21% to 37% compared with the e-GFRs irrespective of race or phase. The studies were undertaken during the daytime and the patients had free access to food and fluid. Increased fluid and osmoles consumption may have positively influenced urine volume and even GFR. Interestingly, the mean difference between CrCl and e-GFR was greater in CCs than AAs, but this failed to achieve statistical significance. The higher e-GFR in African Americans compared with Caucasians is attributable to the MDRD equation since the mean serum creatinine was nearly identical in both groups, and race is a key factor in MDRD equation. The MDRD equation has been suggested to underestimate the prevalence of early-stage chronic kidney disease in African Americans.^32,33 The e-GFRs calculated using CKD-EPI equation were not statistically different from e-GFRs calculated by MDRD equation in this group of patients. This is consistent with previous findings.¹⁴

In conclusion CrCl-12 h was significantly higher than estimated GFR regardless of race in renal allograft recipients. The trend of greater difference between e-GFR and CrCl-12 h in Caucasian than in African American needs to be explored in a larger sample size. Nonetheless, CrCl-12 h seems to provide a reproducible measurement of GFR independent of race. This parameter may be useful for periodic assessment of renal function in RTRs, especially those where the MDRD equation is likely to be more unreliable.

Declaration of interest: The authors have no conflict of interest.

This study was funded in part by an investigator initiated grant from Novartis Pharmaceuticals, East Hanover, NJ.