Assessment of renal function in persons with motor complete spinal cord injury—cystatin C as an accurate single marker

Abstract

Study Design

Observational cohort.

Aim

To show that Cystatin C is an accurate single marker to estimate GFR in motor complete persons with SCI.

Objectives

To assess if Cystatin C is an accurate for estimating GFR in persons with SCI with no preserved motor power. To study if use of Serum creatinine for estimation of GFR in this population significantly overestimates GFR, thereby inaccurate.

Setting

Tertiary care hospital and Medical College, Vellore, South India.

Methods

30 persons with SCI (ASIA A and B) fulfilling the inclusion criteria were recruited. Serum Creatinine and Serum Cystatin C values were obtained, and eGFR was calculated based on available formulae. 24-h urine for urine creatinine clearance-based eGFR was used as a reference value.

Results

Analysis with a Bland-Atman plot showed that eGFR estimated with Serum Cystatin C was more accurate than Serum Creatinine, using 24-h urine creatinine as a reference value. eGFR using Serum Creatinine significantly overestimated GFR by over 50.6%. Estimated GFR using Serum Cystatin C showed a meager mean difference of 0.5% from the reference 24-h urine creatinine clearance (mean difference of –2.56%).

Conclusion

Serum Cystatin C is a much more accurate marker for estimating GFR in SCI, compared to serum Creatinine which overestimates GFR.

Introduction

Individuals with Spinal Cord Injury (SCI) have significant urinary bladder and renal complications. Neurogenic lower urinary tract dysfunction exposes these individuals to various complications, either directly from the dysfunction itself or indirectly from their bladder management protocols.

Individuals with spinal cord injury (SCI) face a risk of developing renal failure or chronic kidney disease (CKD) due to an overactive/hyperreflexic detrusor, which leads to a contracted bladder and the risk of hydro-ureteronephrosis. Recurrent urinary tract infections (UTIs), frequent use of nephrotoxic antibiotics without adequate renal adjustment, NSAID use, and the risk of bladder, ureteric, or renal calculi causing obstructive uropathy are other concerns [1, 2]. UTIs remain the most common infections in SCI patients [3].

Glomerular Filtration Rate estimation (GFR) is a widely accepted method for assessing and monitoring renal function. However, the conventional practice of using serum creatinine levels is questionable in SCI patients. These patients have significantly decreased muscle mass due to atrophy from prolonged disuse in motor-complete SCI patients. As serum creatinine levels depend on muscle mass, they provide an inaccurate measure for estimating GFR, regardless of the formulas used. This underscores the need for an alternative GFR estimation method in this population. Few studies have focused on renal health in SCI patients and even fewer on evaluation methods. To our knowledge, no similar studies have been conducted in India.

Spinal cord injury is known to cause significant sarcopenia. Data on the timeframe of spinal cord injury’s impact on skeletal muscle is scarce. Increased activity might reduce mitochondrial content and enzyme levels due to disuse. Studies on lower mammals indicate a relative independence between metabolic enzyme content in skeletal muscle and activation. Evidence points to significant muscle atrophy within six months of injury, with possibly no further atrophy after two seventeen years [4, 5].

Kidney disease has been identified as a predictor of morbidity in SCI patients. Greenwell et al. demonstrated that proteinuria and a creatinine clearance of less than 60 ml/min independently increased morbidity in SCI patients [6]. The understanding of CKD in SCI is limited by the inaccuracy of current GFR-estimating equations, which significantly overestimate kidney function, potentially leading to the underdiagnosis of CKD in this population [7].

Given these limitations, there is a need for an alternative method to accurately estimate GFR, one not influenced by muscle mass. This led to the adoption of a low molecular mass protein, Cystatin C, originally known as post-gamma-globulin, among other names. The amino acid sequence of human cystatin C was determined in 1981. Unique among cystatins, Cystatin C is produced in all human nucleated cells, indicating stable production due to its housekeeping gene. Large cohort studies have not linked serum levels of Cystatin C to any pathophysiological states other than those affecting GFR [8].

To date, only three studies have assessed the usefulness of Cystatin C in this population, with only two using an accepted gold standard for GFR measurement for comparison [9–11]. The use of 24-h urine creatinine clearance as a GFR estimation method in SCI patients has been recognized, but it is time-consuming, necessitating a faster alternative. This population benefits from more accurate 24-h urine collection compared to the general population [2, 12].

Methodology

Setting

The study was conducted at Christian Medical College (CMC) and Hospital, a tertiary care hospital in Vellore, located in the state of Tamil Nadu, India. Participants were recruited from among the inpatients or those visiting the outpatient services offered by the Department of Physical Medicine and Rehabilitation at CMC.

Study design

This was a prospective observational cohort study. Ethical approval was obtained from the Institutional Review Board to conduct the study. A total of 34 patients who met the inclusion and exclusion criteria were initially recruited; however, 4 patients did not complete the study and were subsequently excluded from the analysis. Thus, data from 30 patients were analyzed.

Demographic data such as age, sex, marital status, mode of injury, and duration of injury were collected from patient records. Measurements required for calculation, such as height and weight, were taken directly. Height was measured either by using the arm span or by measuring the length of a bedridden patient [13].

Participants

Inclusion criteria

1. ASIA A and ASIA B according to American Spinal Injury Association Impairment Scale [14]

2. Age 18 years to 60 years

3. More than 6 months post injury

Exclusion criteria

1. ASIA C and ASIA D

2. Less than 6 months post injury as it may take 6 months for muscle atrophy to set in

3. Known thyroid disease (hypothyroid or hyperthyroid)

4. Age less than 18 years or more than 60 years (may be associated with age related sarcopenia)

5. Pregnancy and lactation

Only ASIA A and ASIA B persons with SCI were recruited [14]. This was to ascertain that complete motor injury patients were recruited. This thereby maintaining that the overestimation of Serum Creatinine estimated GFR would be because of significant muscle atrophy which may not be elicited if motor incomplete SCI patients are included in the study.

ASIA A was defined as—complete—no sensory or motor function is preserved in the sacral segments S4-5.

ASIA B was defined as—sensory incomplete—Sensory but not motor function is preserved below the neurological level and includes the sacral segments S4-5 (light touch or pin prick at S4-5 or deep anal pressure) AND no motor function is preserved more than three levels below the motor level on either side of the body.

Blood samples were tested for the following:

1. Serum Creatinine (mg/dL)

2. Cystatin C (mg/L)

3. Albumin (g/dL)

estimated GFR (eGFR) was calculated using the following formulae:

Creatinine eGFR:

1. Cockcroft Gault GFR (with Body wt.)—(ml/min/1.73m2)—without using the patient’s body surface area. (Creat GFR1)

2. Cockcroft Gault GFR (with Body wt.)—(ml/min)—using the height and weight of the patient to derive body surface area. (Creat GFR2)

3. CKD–EPI creatinine—(ml/min/1.73m2) (Creat GFR3)

Cystatin C eGFR:

1. CKD–EPI—(ml/min/1.73m2)—without using the patient’s body surface area. (Cys GFR1)

2. CKD–EPI—(ml/min)—using the height and weight of the patient to derive body surface area. (Cys GFR2)

3. Combined Cystatin C and Creatinine eGFR – ml/min/1.73m2 (Comb GFR).

24—hour urine creatinine clearance

24—h urine samples were used to calculate urine creatinine clearance.

GFR was calculated from 24-h urine collection by the formula [urine concentration] x (urine flow rate) / [plasma concentration], or simply, C = UV/P. This value was used as a reference value for our study.

Urine sample for UP/UC ratio was collected.

Serum creatinine samples were collected after urine collection.

Ten patients who underwent renograms underwent urine collection prior to the renograms to prevent radionuclide contamination.

This method was used as a reference for comparison of GFR.

Athough levels of Serum creatinine in SCI patients are expected to lower than normal due to sarcopenia (muscle atrophy), its renal clearance measured as urine creatinine levels would remain a marker of GFR when compared their serum levels. (UV/P)

Technitium 99 -Tc ^99m DTPA scan using gamma camera imaging was used in 10 subjects to calculate eGFR using Gate’s method. There was no provision to use serial plasma clearance of a radionucleotide for this study.

We performed this test only in 10 patients due to marked variability in the results, high cost and the fact that it was a time-consuming test. It was not validated in our population of study and, hence the remainder of the subjects did not undergo the test.

Although an ideal gold standard (serial plasma clearance of 51Cr -EDTA or 99Tc – DTP) was unavailable in our study for comparison, 24-h urine creatinine clearance is accepted as a good reference value.

Results

Sample size calculation

Using the results from a previous study, the estimated GFR (eGFR) calculated from Cystatin C mean (SD) reported was 83.7(± 19.8). 90% measured eGFR was conferred by eGFR (Cystatin C ± 30%). Witha 30% difference from Cystatin C eGFR, the difference was around 25 units (ml/min).

Statistical calculation estimated that a sample size of 15 subjects was sufficient to detect a mean difference of 20 units considering a standard deviation (SD) of 25 units, with a power of 80% and a significance level of 5% (Eq. (1))

$$\begin{array}{c}{N}_{pairs}=\frac{{\left(Z_{1-\alpha /2}+Z_{1-\beta }\right)}^2}{{\Delta }^2}+\frac{Z_{1-\alpha /2}^2}{2}\\ \Delta =\frac{{\mu }_2-{\mu }_1}{\sigma }\sigma =\frac{{\sigma }_1+{\sigma }_2}{2}\end{array}$$ (1)

Where,µ₁ = mean of Cystatin Cµ₂ = Mean of measured eGFRσ₁ = Standard deviation of Cystatin Cσ₂ = Standard deviation of measured eGFRZ_1-α/2 = Desired confidence levelZ _1-β = Power.

A total of 34 patients were recruited of whom 30 patients completed the evaluation.

Hence data for 30 patients over a period of 1 year was available.

The mean age of the participants was 33.8 ± SD 9.43, (18-54 years).

There were 29 male and 1 female patients. Six patients had tetraplegia and 24 had paraplegia.

25 had ASIA A and 5 had ASIA B SCI.

40% of patients had an indwelling urethral catheter, 50% practiced Self Intermitted Clean Catheterisation and 10% of patients had a suprapubic indwelling catheter.

(15)Formulae that were used for eGFR calculation were a follows

• –Cr GFR1–CKD–EPI creatinine – (ml/min/1.73m2)
• –Cr GFR2- Creatinine-Cockcroft Gault GFR (with Body wt.) – (ml/min/1.73m2)
• –Creat GFR3 – Creatinine Cockcroft Gault GFR (with Body wt.) – (ml/min) -using body surface area.
• –Cys GFR1–CKD–EPI Cystatin C - (ml/min/1.73m2)
• –Cys GFR2 - CKD–EPI Cystatin C- (ml/min) – using body surface area.
• –Comb GFR–Combined Cystatin C and Creatinine eGFR – ml/min/1.73m2

Bland Altman method for analysis

The Bland Altman method. A modification of this method in 2008 by Krouwer was used for comparison [15].

Comparison between the percentage mean difference between GFR estimated by each separate method Vs the reference GFR (which is 24-h urine creatinine clearance) was analysed.

Percentage mean difference was calculated by:

$$\frac{\text{EstimatedGFR}\text{(by}\text{thevarious}\text{methods})-\text{reference}\text{GFR}}{\text{reference}\text{GFR}}\times 100$$

This percentage mean difference was plotted against the Y – Axis and the Reference GFR value was plotted on the X – axis. Overestimation or underestimation of estimated GFR values could be observed at different GFR levels.

The arithmetic mean and the 95% confidence intervals (CI) for the upper and lower limits were calculated.

The CKD–EPI folmula overestimates GFR by a mean of 50.6%.

(Figure 1)

Fig. 1. Cr GFR1: CKD–EPI creatinine – (ml/min/1.73m2).

Correlation of CKD - EPI Creatinine equation to 24 urine Creatinine Clearance based GFR.

(Table 1)

Table 1. Cr GFR1: CKD–EPI creatinine—(ml/min/1.73m2) Reference GFR: 24-h urine creatinine clearance (ml/min).

Mean	50.6141
95% CI	36.8810 to 64.3473
P (H0: Mean=0)	<0.0001
Lower limit	−21.4709
95% CI	−45.2060 to 2.2642
Upper limit	122.6992
95% CI	98.9641 to 146.4342

The Creatinine based Cockcroft Gault formula overestimates GFR by a mean of 45.09%.

(Figure 2)

Fig. 2. CrGFR 2: Creatinine- Cockcroft Gault GFR (with Body wt.) – (ml/min/1.73m2).

Correlation of Creatinine base Cockroft Gault formula with 24 hour urine creatinine clearance based GFR.

(Table 2)

Table 2. CrGFR 2: Creatinine-Cockcroft Gault GFR (with Body wt.) – (ml/min/1.73m2) Reference GFR: 24-h urine creatinine clearance (ml/min).

mean	45.0881
95% CI	33.5376 to 56.6386
P (H0: Mean=0)	<0.0001
Lower limit	−15.5400
95% CI	−35.5028 to 4.4227
Upper limit	105.7162
95% CI	85.7535 to 125.6789

The Cystatin C based CKD – EPI formula remains close to the reference GFR by a mean difference of –0.47%. The p value is closer to one, indicating the null hypothesis that there isn’t much significant difference between the 2 methods.

(Figure 3)

Fig. 3. CysGFR 1: CKD–EPI Cystatin C - (ml/min/1.73m2).

Correlation of CKD - EPI Cystatin C equation to 24 urine Creatinine Clearance based GFR.

(Table 3)

Table 3. CysGFR 1: CKD–EPI Cystatin C - (ml/min/1.73m2) Reference GFR: 24-h urine creatinine clearance (ml/min).

Mean	−0.4727
95% CI	−9.1373 to 8.1918
P (H0: Mean=0)	0.9119
Lower limit	−45.9529
95% CI	−60.9279 to −30.9778
Upper limit	45.0074
95% CI	30.0324 to 59.9824

Comparison between combined cystatin C and creatinine CKD—EPI formula VS reference GFR

The combined GFR (Using cystatin C and creatinine, CKD–EPI) overestimated GFR by 19.76%.

(Table 4)

Table 4. Combined GFR—Combined cystatin C and creatinine eGFR(CKD–EPI) – ml/min/1.73m2 Reference GFR: 24-h urine creatinine clearance (ml/min).

mean	19.7676
95% CI	11.5359 to 27.9993
P (H0: Mean=0)	<0.0001
Lower limit	−23.4405
95% CI	−37.6675 to −9.2136
Upper limit	62.9757
95% CI	48.7488 to 77.2027

Based on CKD stages:

Normal GFR—11 patients (>90 ml/min) 36.67%

Stage 2—13 patients (60–89 ml/min) 43.33%

Stage 3–4 patients (30–59 ml/min) 13.33%

Stage 4—2 patients (15–29 ml/min) 6.67%

63.33% of the patients had GFR below 90 ml/min (normal).

Discussion

Creatinine-Based Equations Significantly Overestimate GFR

The Bland-Altman plot method unequivocally demonstrated the overestimation of GFR by serum creatinine-derived equations. The extent of overestimation is highlighted by the large discrepancies (50.6%, 45.1%, and 38.3% for three different serum creatinine-based equations) from the reference GFR value, all of which exhibited significant P-values (<0.0001). These overestimations render serum creatinine an unreliable marker for clinical practice.

The newly developed creatinine-based CKD–EPI formula exhibited the largest mean difference, indicating substantial over-estimation across all methods tested. The combined Cystatin C and creatinine-based CKD–EPI equation also revealed a 19.6% mean overestimation, which, while lower than that of the creatinine-based equations, was still significantly high.

Cystatin C as an Effective Single Marker for Estimating GFR in SCI

The differences between the mean differences of the Cystatin C-based CKD–EPI equation GFRs and the reference GFR were minimal (-0.5% and -2.6%), We found, serum Cystatin C to be an accurate alternative marker for GFR estimation in spinal cord injury patients with ASIA A and B scores. Serum Cystatin C levels can be quickly determined obviating the need for 24-h urine collection. Moreover the equations for GFR estimation (Cystatin C-based CKD–EPI equations) which we used did not require body surface area estimation, which is cumbersome in the SCI patient.

Incidental Finding of Renal Failure in Spinal Cord Injury Patients

A significant proportion of patients (63.33%) had GFR values below the normal range (90 ml/min), indicating a substantial burden of renal failure in this group of patients. Additionally, 20% were classified into Stages 3 and 4 of CKD, necessitating management and periodic follow-up. Two patients had a GFR of <30 ml/min. Hence these patients required nephrological consultation for further management and evaluation.

Other factors contributing to an increased risk of CKD in this population included low socioeconomic status, inadequate diet, recurrent infections, and limited access to healthcare, which are particularly relevant in developing countriesThe impact of long-term bladder management methods on renal health in SCI patients also requires further investigation.

Conclusion

Cystatin C is an accurate single marker for GFR estimation in individuals with motor complete SCI. Using serum creatinine to estimate GFR in these patients significantly overestimates GFR and is therefore not recommended for clinical practice. The high prevalence of compromised renal function in spinal cord injury patients highlights the need for more rigorous evaluation and screening of renal health in this population. Further studies are necessary to assess renal health comprehensively in this population.

Data Availability

Data analysed is present in the article. Additional data are available from the corresponding author on reasonable request.