Original Article KIDNEY RESEARCH

Kidney Res Clin Pract 37:338-346, 2018(4)

pISSN: 2211-9132 • eISSN: 2211-9140
AND
CLINICAL PRACTICE
https://doi.org/10.23876/j.krcp.18.0080

Cystatin C as a novel predictor of preterm labor in

severe preeclampsia
Krittanont Wattanavaekin1, Maethaphan Kitporntheranunt2, Chatchai Kreepala3
1
Faculty of Medicine, Srinakharinwirot University, Nakhon Nayok, Thailand
2
Department of Obstetrics and Gynecology, Faculty of Medicine, Srinakharinwirot University, Nakhon Nayok, Thailand
3
Department of Internal Medicine, Faculty of Medicine, Srinakharinwirot University, Nakhon Nayok, Thailand

Background: The most common cause of acute kidney injury (AKI) in pregnancy is preeclampsia. Serum cystatin C
(CysC) is a potential biomarker of early kidney damage as its levels are not disturbed by volume status changes in
pregnancy, and serum CysC levels could serve as a replacement for conventionally used creatinine. In this study, we
investigated the serum levels of CysC in severe preeclampsia cases and the associations between CysC levels and
poor obstetric outcomes.
Methods: Our cohort included severe preeclampsia patients with a normal serum creatinine level. Creatinine was
measured to calculate estimated glomerular filtration rate (eGFR) based on the Cockcroft and Gault, Modification of
Diet in Renal Disease Study (MDRD), and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations,
while CysC was measured to calculated eGFR based on a CysC-based equation. We then evaluated the correlations
between serum CysC level, eGFR, and obstetric outcomes.
Results: Twenty-six patients were evaluated of which 38.5% delivered preterm and 30.8% had low-birth weight
babies. Unlike creatinine-based eGFR and CysC-based eGFR, serum CysC demonstrate significant negative
correlation with gestational age. Receiver operating characteristic curve analysis indicated that serum CysC is
a potential biomarker of preterm delivery with a cut-off serum level of 1.48 mg/L with 80% sensitivity and 75%
specificity.
Conclusion: GFR estimation using CysC is likely to be inaccurate in pregnancy. However, we found a significant
correlation between preterm delivery and serum CysC level. Our results suggest that serum CysC level has the
potential to predict preterm delivery in severe preeclampsia patients.

Keywords: Acute kidney injury, Cystatin C, Low birth weight infant, Preeclampsia, Premature obstetric labor

Introduction

Received July 25, 2018; Revised August 23, 2018; Preeclampsia is a pregnancy-specific disorder that af-
Accepted August 29, 2018 fects 4% of pregnancies and is routinely diagnosed when
Correspondence: Chatchai Kreepala
Department of Internal Medicine, Faculty of Medicine, Srinakharinwirot
a pregnant woman presents with increased blood pres-
University, 62 Moo 7, Her Royal Highness Princess Maha Chakri sure and proteinuria [1]. Preeclampsia is one of the main
Sirindhorn Medical Center, Ongkharak District, Nakornnayok Province
26120, Thailand. E-mail: [email protected]
causes of maternal, fetal, and neonatal mortality, espe-
ORCID: https://orcid.org/0000-0002-2615-4222 cially in developing countries [2]. The acute clinical im-
Copyright © 2018 by The Korean Society of Nephrology portance of preeclampsia lies in its relation to maternal
This is an open-access article distributed under the terms of the Creative Commons
and neonatal mortality and morbidity. Pregnant women
CC

Attribution Non-Commercial License (http://creativecommons.org/licenses/by-

nc-nd/4.0/), which permits unrestricted non-commercial use, distribution, and with preeclampsia can experience severe complications
reproduction in any medium, provided the original work is properly cited.
such as eclampsia, HELLP syndrome, pulmonary edema,
 Wattanavaekin, et al. CysC, a biomarker of AKI, as a predictor of preterm labor

or kidney failure [3]. The disease is also related to fetal superfamily of cysteine protease inhibitors that play im-
growth restriction and preterm delivery. An increase in portant roles in intra-cellular catabolism of proteins and
blood pressure can develop into renal dysfunction, and peptides. Some studies have shown that this biomarker
worsening of renal function will result in salt and water is a better indicator of kidney function than creatinine
retention, which further complicate the hypertensive [17,18], but there are no widespread recommendations
state. This vicious cycle can ultimately lead to severe pre- for its use, even in pregnancy [19]. More intensive studies
eclampsia, followed by poor obstetric outcome. Children will have to be conducted before it can become a rou-
born to mothers with preeclampsia have an increased tinely used clinical biomarker. Several researches have
risk of bronchopulmonary dysplasia and cerebral palsy described the utility of CysC as a biomarker in pregnancy.
caused by preterm birth and being small for gestational However, no prior study has investigated whether this
age [4,5]. biomarker is a predictor of poor obstetric outcomes [20-
The most common causes of acute kidney injury (AKI) 24].
in pregnancy are preeclampsia and eclampsia, which can In this study, we used serum creatinine and serum CysC
ultimately lead to an increase in both fetal and maternal to calculate eGFR using various formulae, and investi-
mortality rates [6]. Many biomarkers are available for the gated if serum CysC level is associated with the obstetric
detection of kidney damage, and early AKI detection may complications of preterm delivery and low-weight babies
be a potential predictor of preeclampsia [7-11]. in patients with severe preeclampsia.
Chapman et al [12] documented early rises in glomeru-
lar filtration rate (GFR) and kidney blood flow by inulin Methods
and p-aminohippurate clearance in association with sys-
temic and kidney vasodilation in a series of 10 pregnant Clinical data collection
women [12]. However, as inulin and p-aminohippurate
levels are determined by invasive investigation, these We enrolled a prospective cohort of pregnant women
biomarkers are not suitable for screening for AKI in the diagnosed with severe preeclampsia who were receiv-
setting of preeclampsia. Serum creatinine, which is wide- ing antenatal care at Her Royal Highness Princess Maha
ly used as a marker of kidney damage, has been shown to Chakri Sirindhorn Medical Center (MSMC), Thailand
be a poor biomarker of early AKI in pregnancy because of during August 1st, 2014 and September 30th, 2016. All
its low sensitivity [13]. Kidney Disease Improving Global women voluntarily signed a consent form before partici-
Outcomes (KDIGO) specified that normal serum creati- pating in our study. Data were collected from pregnant
nine levels are 0.4 to 0.8 rather than 0.8 to 1.2 in preg- women with a maternal age more than 18 years and in the
nancy due to underestimation caused by dynamic vol- second half of pregnancy (gestational age, > 20 weeks).
ume status. There are also no available creatinine-based However, we included only severe preeclampsia cases
formulae to accurately calculate estimated GFR (eGFR) with a gestational age of more than 28 weeks in our co-
in pregnant patients using serum creatinine [14]. Several hort. Those with a history or diagnosis of chronic hyper-
studies have also shown that serum creatinine is unreli- tension, placenta previa, gestational diabetes, or chronic
able compared to inulin, even though serum creatinine kidney disease with serum creatinine levels ≥ 0.8 mg/dL
is the gold standard for calculation of GFR. Kidney injury were excluded. Sample size was calculated using a power
molecule-1 and neutrophil gelatinase-associated lipo- of 80%, alpha value of 0.05, and a dichotomous endpoint
calin are other promising biomarkers of AKI. However, formula. Using data from a pilot project involving 10 pre-
because the levels of these markers increase rapidly and eclamptic patients (unpublished data), we found that the
normalize within 48 hours, they are not appropriate for sensitivity and specificity of serum CysC for predicting
use as screening biomarkers [15]. preterm deliveries was 0.75 and 0.60, respectively. Using
Cystatin C (CysC) has been studied intensively in preg- the Buderer approach and a severe preeclampsia preva-
nant women because it is independent of body weight, lence of 1.3%, we calculated that we needed a minimum
muscle mass, and other pregnancy-related changes [16]. sample size of 25 [25,26].
CysC is a 13 kDa proteinase inhibitor from the cystatin A clinical study registration number for this study is

www.krcp-ksn.org 339
Kidney Res Clin Pract Vol. 37, No. 4, December 2018

TCTR20180730002. The protocol for patient participation using the Chronic Kidney Disease Epidemiology Collabo-
was approved by the Human Research Ethics Board of ration (CKD-EPI) CysC equation (2012) [33]. The eGFR
Srinakharinwirot University (Issue #SWUEC/E-038/2557), based on creatinine was calculated using the CKD-EPI
and fulfilled the tenants of the 1964 Helsinki declaration creatinine equation (2009) [34], Modification of Diet in
and its amendments. Renal Disease Study (MDRD) creatinine equation (2006)
[35], and Cockcroft and Gault creatinine equation. In-
Clinical assessment of obstetric outcomes dividual eGFR was assessed using a web-based calcula-
tor (https://www.kidney.org/professionals/kdoqi/gfr_
Severe preeclampsia was defined as a systolic blood calculator) [36] after entering serum creatinine or CysC,
pressure (SBP) ≥ 160 mmHg or a diastolic blood pressure age, and standardization assay without body surface
(DBP) ≥ 110 mmHg on two occasions at least 4 hours apart adjustment. The Cockcroft and Gault creatinine equation
in a previously normotensive patient with a positive urine (eGFR-CG) [37] was expressed as follows:
protein result (at least 1+ of dipstick urine protein), im- eGFR = {[(140 - age) × weight (kg)]/serum creatinine ×
paired hepatic function, progressive renal insufficiency, or 72} × 0.85.
new onset cerebral/visual disturbance, pulmonary edema, Urine protein level was assessed semi-quantitatively
or thrombocytopenia (platelet count, < 100,000/mm3) [27]. using the colorimetric method of tetrabromophenol
Preterm labor was defined as babies born alive before 37 blue reagent in urine dipstick [38]. The threshold used
weeks of pregnancy [28] and low birth weight as a weight to define a positive test result was 300 mg/24 hours, 300
at birth of less than 2,500 g (5.5 pounds) or a fetus in the mg/dL, or a dipstick proteinuria reading of more than 1+
lowest 10th percentile of the fetal growth chart (gestation- defined based on the diagnostic criteria of preeclampsia
al age confirmed by both date and ultrasound in all pa- [39]. Only patients diagnosed with preeclampsia based
tients) [29,30]. Teenage pregnancy was defined as births on hypertension and persistently high thresholds (> 1+
to mothers aged 15 to 19 years and elderly pregnancy was proteinuria) were included in the study.
defined as birth by mothers older than 35 [31,32].
Patients with severe preeclampsia received a magne- Statistical analyses
sium infusion based on MSMC guidelines if their ges-
tational age was more than 28 weeks and the obstetri- IBM SPSS statistical software package ver. 23.0 (IBM
cian had ensured lung maturity of the neonate. Within Corp., Armonk, NY, USA) was used for all statistical
24 hours after infusion of magnesium, termination was analyses. Continuous data were evaluated to determine
considered. All patients had blood samples taken for in- whether they followed a normal distribution. Parametric
vestigation of complete blood count, liver function tests, data were evaluated using parametric tests and Pear-
and creatinine and CysC level determination for initial son correlation coefficients. Non-parametric data were
admission before receiving a magnesium infusion. Clini- evaluated using non-parametric tests and Spearman’s
cal parameters such as blood pressure, volume of fluid coefficient was calculated to assess the associations be-
intake, urine output, and blood loss were monitored rou- tween various parameters and obstetric outcomes. Mean
tinely until end of delivery. values and standard deviations were compared among
groups using Student’s t test. Pearson correlation and
Laboratory evaluation logistic regression analyses were used to determine the
significance of correlations between the various clinical
Serum creatinine was measured by the enzymatic parameters and obstetric outcomes. The area under the
method using a Cobas® 6000 analyzer (Roche Diagnos- receiver operating characteristic (ROC) curve was used
tics, Mannheim, Germany). Serum CysC was analyzed by to assess diagnostic accuracy and to perform cross-vali-
an immunoturbidimetric method using an ARCHITECT dation analysis. P values less than 0.05 were considered
ci 8200 instrument (Abbott Laboratories Ltd., Bangkok, statistically significant.
Thailand). All formulae were adjusted for non-black fe-
male patients. The eGFR based on CysC was calculated

340 www.krcp-ksn.org
 Wattanavaekin, et al. CysC, a biomarker of AKI, as a predictor of preterm labor

Results weight corrected by gestational age was 19.2% in this co-

hort (Table 1). Other complications often associated with
Patient characteristics preeclampsia like hepatitis or HELLP were not found in
the present study. Other demographic data are presented
Twenty-six pregnant women with severe preeclampsia in Table 1.
who delivered at MSMC were analyzed in the present
study. All patients had a normal serum creatinine level Association between eGFR calculated using various
(0.40 to 0.80 mg/dL), and serum CysC ranged from 0.99 to creatinine or CysC-based formulae
1.86 mg/L (1.43 ± 0.24 mg/L, mean ± standard deviation).
Ten (38.5%) women who experienced preeclampsia were Four formulae were used to calculated eGFR: the CKD-
teenage or elderly pregnancies. The average gestational EPI CysC equation (2012), CKD-EPI creatinine equation
age at delivery was 37.07 ± 2.30 weeks. The incidence of (2009), MDRD equation (2006), and Cockcroft and Gault
preterm delivery was 38.5% (gestational age was con- equation. A significant directional correlation was found
firmed by date and ultrasound). Incidence of low birth among all eGFR values. However, GFR estimated using
the CysC-based calculation was drastically different from
that estimated using the creatinine-based equations, with
Table 1. Characteristics of pregnant women with severe the CysC-based eGFR less than half of the creatinine-
preeclampsia based eGFR values (Table 2, 3). Additionally, we found
Clinical variable Value that up to 46.2% of severe pre-eclamptic patients had
Subject (n) 26 an eGFR < 60 mL/minute when calculated with CysC,
Maternal age (yr) 29.9 ± 7.6 whereas their serum creatinine levels were in the normal
Teenage pregnancy* 3 (11.5) range. This implies that the creatinine-based equation
Elderly pregnancy† 7 (26.9) may overestimate the real GFR, or that the CysC-based
Systolic blood pressure (mmHg) 167.5 ± 12.1 (160-220) formula may underestimate the GFR. If the former is cor-
Diastolic blood pressure (mmHg) 104.9± 11.2 (90-140) rect, CysC may potentially be more sensitive as an early
Gestational age at delivery (wk) 37.1 ± 2.3 AKI biomarker than creatinine.
Preterm delivery 10 (38.5)
Low birth weight‡ 5 (19.2)
CysC-based GFR and clinical outcomes
Placental weight (g) 571.5 ± 149.4 (371.1-913.0)
Serum creatinine (mg/dL) 0.6 ± 0.1 (0.40-0.79)
Binary logistic regression analysis was employed to
Serum cystatin C (mg/L) 1.4 ± 0.2 (0.99-1.86)
predict risk of the adverse obstetric outcomes of preterm
Hematologic profile
Hemoglobin (g/dL) 12.0 ± 1.2 delivery and low birth weight body adjusted for maternal
Platelets (×103/mm3) 275.1 ± 92.4
LDH (U/L) (normal, 125-220) 204.3 ± 65.3
PT (sec) (normal, 12-14.6) 11.9 ± 0.8 Table 2. Correlation between glomerular filtration rate (GFR)
INR (normal, 0.8-1.2) 1.0 ± 0.1 estimated using different biomarkers and equations
aPTT (sec) (normal, 25.1-35.1) 27.8 ± 3.0 Equation
Liver function tests   Cockcroft
CKD-EPI MDRD CysC
AST (U/L) (normal, 5-34) 19.9 ± 7.9 and Gault
ALT (U/L) (normal, 0-54) 14.0 ± 7.4 CKD-EPI creatinine 0.91** 0.75** 0.41*
at time 0
Data are presented as number only, mean ± standard devation (range), or
number (%).
MDRD at time 0 0.91** 0.81** 0.52**
ALT, alanine aminotransferase; aPTT, activated partial thromboplastin time; Cockcroft and Gault 0.75** 0.81** 0.45*
AST, aspartate transaminase; INR, international normalized ratio; LDH, lactate at time 0
dehydrogenase; PT, prothrombin time. CysC-GFR 0.41* 0.52** 0.45*
*Age ≤ 19 years, †age ≥ 35 years. CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; CysC-GFR,
‡
Low birth weight corrected by gestational age of a fetal-growth chart developed cystatin C-based GFR; MDRD, Modification of Diet in Renal Disease Study.
for Thai pregnant women (Kiserud et al [30]). *P < 0.05, **P < 0.01.

www.krcp-ksn.org 341
Kidney Res Clin Pract Vol. 37, No. 4, December 2018

Table 3. Average individual eGFR calculated using various equations

eGFR equation Type Data (mL/min)
CKD-EPI cystatin C equation (2012) CysC-based 53.2 ± 12.8 (48.3-58.1)
CKD-EPI creatinine equation (2009) Creatinine-based 125.2 ± 11.6 (120.7-129.6)
MDRD equation (2006) Creatinine-based 133.0 ± 28.3 (122.1-143.9)
Cockcroft and Gault equation Creatinine-based 186.7 ± 47.0 (168.6-204.7)
Data are presented as mean ± standard deviation (95% confidence interval).
CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; CysC, cystatin C; eGFR, estimated glomerular filtration rate; MDRD, Modification of Diet in Renal
Disease Study.

Table 4. Binary logistic regression analysis for predicting ROC curve

1.0
preterm delivery and low birth weight adjusted by maternal
age and systolic and diastolic blood pressure
0.8
Predicting Predicting a low
Variable preterm delivery birth weight baby

Sensitivity
0.6
OR P value OR P value
Serum cystatin level (mg/L) 32.52 < 0.01 1.08 0.3
0.4
Serum creatinine level 77.87 0.32 6.94 0.2
(mg/dL)
0.2
eGFR based on
CysC (mL/min) 1.06 0.13 1 0.29 Area under ROC curve = 0.738
P value = 0.045
0.0
CKD-EPI creatinine (mL/min) 1.07 0.96 1.01 0.27 0.0 0.2 0.4 0.6 0.8 1.0
MDRD equation (mL/min) 1.01 0.77 1.01 0.27 1-Specificity
Cockcroft and Gault 1.00 0.74 1.00 0.28 Cut-off level of CysC (mg/L) Sensitivity Specificity
equation (mL/min) 1.06 0.9 0.06
CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; CysC, cystatin C; 1.28 0.9 0.44
eGFR, estimated glomerular filtration rate; MDRD, Modification of Diet in Renal 1.48 0.8 0.75
Disease Study; OR, odds ratio. 1.54 0.6 0.75
1.68 0.3 0.88

age and systolic and DBP. Only serum CysC was a signifi- Figure 1. Receiver operating characteristic (ROC) curve of the
cant predictor of preterm delivery (odds ratio, 32.52; P < predictive accuracy of cystatin C (CysC) for preterm delivery in
preeclamptic patients. The area under the ROC curve was 0.74,
0.01). The eGFR calculated by creatinine- or CysC-based
indicating a significant degree of discrimination (P = 0.04). CysC
formulae did not predict preterm delivery. Neither renal values > 1.48 mg/L were highly suggestive of preterm delivery in se-
function biomarkers (creatinine and CysC) nor eGFR vere preeclamptic patients with 80% sensitivity and 75% specificity.
could predict a low birth weight (Table 4). Proteinuria
(protein dipstick) and systolic and DBP were also not screening for the risk of preterm delivery in patients with
predictive of low birth weight or preterm delivery. preeclampsia with a sensitivity and specificity of 80% and
75%, respectively. Thus, a pregnant woman with in pre-
Accuracy of serum CysC for predicting preterm birth in eclampsia with a serum CysC level > 1.48 mg/L would be
preeclampsia patients considered at risk for preterm delivery and require close
monitoring and constant medical attention (Fig. 1).
We further used ROC curve analysis to evaluate the ac-
curacy of serum CysC for predicting preterm birth in our Discussion
preeclampsia cohort. Analysis showed that the area un-
der the curve was 0.738 with P = 0.045, indicating a high Renal function is often closely monitored in pre-
degree of discrimination. eclamptic patients as it is a major concern [40,41] with
ROC curve analysis suggested that the cut-off serum the primary goal of timely delivery before irreversible
CysC level of > 1.48 mg/L was the best threshold for kidney injury develops. Serum creatinine, serum CysC,

342 www.krcp-ksn.org
 Wattanavaekin, et al. CysC, a biomarker of AKI, as a predictor of preterm labor

and uric acid concentration are parameters that have cohort of over a thousand individuals who had a history
been used previously to monitor renal function in hyper- of coronary heart disease. Creatinine and estimated cre-
tensive disorders of pregnancy [42-45]. atinine clearance were not significantly associated with
Hemodynamic adaptations in pregnant women is cru- risk of a cardiovascular event. In contrast, higher plasma
cial for healthy pregnancy outcomes [46]. These adapta- CysC was associated with an increased risk of a CV event,
tions include a reduction in vascular resistance, increase even after adjusting for well-known risk factors, includ-
in cardiac output, and increase in intravascular volume ing C-reactive protein. Compared with individuals in the
via sodium and water retention [47]. These changes also lowest quintile of CysC, those in the highest quintile had
affect other organs. For instance, there will be an increase a more than a two-fold increase in risk, even after adjust-
in GFR and effective renal plasma flow within the kidney, ing for estimated creatinine clearance [59]. In this study,
which are essential for maintenance of electrolyte bal- we found a correlation between serum CysC level and
ance [48]. Poor pregnancy outcomes are often observed preterm delivery. Serum creatinine, semi-quantitative
in pregnancies where these physiological changes have urine protein level, and hypertension showed no ability
failed to occur [49]. A study by Piccoli et al [50] in 2015 to predict obstetric outcomes in patients with severe pre-
demonstrated a worse prognosis in mothers with stage 1 eclampsia.
CKD. In normal pregnancy, serum CysC levels are high (0.89 ±
CysC is produced by nucleated cells at a constant rate, 0.12 mg/L) in the first trimester, decrease significantly (0.65 ±
and is then filtered and excreted by nephrons. Despite 0.14 mg/L) during the second trimester (P < 0.001 compared
early enthusiasm, however, CysC has not been shown to to first trimester), and increase again (0.82 ± 0.19 mg/L) in
have obvious superiority compared to creatinine [51]. the third trimester.
The reason for this is that even though CysC levels are In this study, the average serum CysC levels during
relatively constant in individuals under normal circum- the third trimester in pre-eclamptic patients was 1.43 ±
stances, it can be affected by several factors including 0.24 mg/L (Table 1), which is significantly higher than
smoking, liver disease, thyroid disease, diabetes mellitus, in normal pregnancy. This suggests that CysC may have
extreme age, and obesity [52-57]. Additionally, the high- adequate sensitivity to detect abnormal renal function
er cost associated with measuring CysC and lack of avail- even when serum creatinine levels are within the normal
ability of tests has hindered its acceptance as a creatinine range. This has clinical significance because even small
replacement to estimate renal function, even in pregnant reductions in renal function are associated with adverse
patients [58]. pathophysiologic consequences. By plotting a ROC curve,
We found that GFR values estimated using both cre- we found that a serum CysC concentration of higher than
atinine and CysC were directionally and proportionally 1.48 mg/L has a sensitivity of 80% and specificity of 75%
correlated. However, CysC-based GFR was approximately to detect preterm delivery in patients with preeclampsia
half the value of the creatinine-based eGFR estimates, (Fig. 1).
suggesting that CysC underestimates the true GFR value. In conclusion, although CysC is a biomarker for detec-
This also indicates that there are uninvestigated compo- tion of early AKI, eGFR using CysC during pregnancy is
nents in the CysC formula that makes it unsuitable for unlikely to be accurate, suggesting unexplored inflam-
accurate estimation of GFR in pregnant women. Inter- matory processes that cause preterm delivery in mothers
estingly, none of the eGFR values were correlated with a with preeclampsia. Our results indicate that CysC is a val-
poor obstetric outcome. id inflammatory marker that can predict preterm delivery
CysC may nevertheless have prognostic importance. in addition to being a biomarker of AKI. Serum CysC level
CysC, in addition to being a biomarker of AKI, is also a therefore has the potential to predict preterm delivery in
known marker of inflammation [58]. There is mounting patients with severe preeclampsia.
evidence that CysC may be a predictor of adverse out-
comes independent of renal function. A study by Koenig Conflicts of interest
et al [59] examined the association between plasma CysC
and risk of secondary cardiovascular (CV) events in a All authors have no conflicts of interest to declare.

www.krcp-ksn.org 343
Kidney Res Clin Pract Vol. 37, No. 4, December 2018

Acknowledgments [10] Kreepala C, Luangphiphat W, Villarroel A, Kitpornther-

anunt M, Wattanavaekin K, Piyajarawong T. Effect of mag-
The authors would like to express our appreciation for a nesium on glomerular filtration rate and recovery of hy-
grant from the Her Royal Highness Princess Maha Chakri pertension in women with severe preeclampsia. Nephron
Sirindhorn Medical Center (MSMC), Thailand. We grate- 138:35-41, 2018
fully acknowledge use of the services and facilities of Her [11] Kreepala C, Kitporntheranunt M, Sangwipasnapaporn W,
Royal Highness Princess Maha Chakri Sirindhorn Medi- Rungsrithananon W, Wattanavaekin K. Assessment of pre-
cal Center (MSMC), funded by MSMC Grant contract eclampsia risk by use of serum ionized magnesium-based
number 461/2557. We also would like to express our cor- equation.  Ren Fail 40:99-106, 2018
dial thanks to the staff of the Department of Pathology, [12] Chapman AB, Abraham WT, Zamudio S, et al. Temporal re-
the Antenatal Care Unit, and the Labour Unit of the De- lationships between hormonal and hemodynamic changes
partment of Obstetrics and Gynecology, Faculty of Medi- in early human pregnancy.  Kidney Int 54:2056-2063, 1998
cine, Srinakharinwirot University for clinical data review. [13] Huang C, Chen S. Acute kidney injury during pregnancy and
puerperium: a retrospective study in a single center. BMC
References Nephrol 18:146, 2017
[14] Smith MC, Moran P, Ward MK, Davison JM. Assessment
[1] Ananth CV, Keyes KM, Wapner RJ. Pre-eclampsia rates in the of glomerular filtration rate during pregnancy using the
United States, 1980-2010: age-period-cohort analysis. BMJ MDRD formula. BJOG 115:109-112, 2008
347:f6564, 2013 [15] Siew ED, Ware LB, Ikizler TA. Biological markers of acute
[2] Saleem S, McClure EM, Goudar SS, et al. A prospective kidney injury.  J Am Soc Nephrol 22:810-820, 2011
study of maternal, fetal and neonatal deaths in low- and [16] Filler G, Bökenkamp A, Hofmann W, Le Bricon T, Martínez-
middle-income countries. Bull World Health Organ 1:605- Brú C, Grubb A. Cystatin C as a marker of GFR—history,
612, 2014 indications, and future research.  Clin Biochem 38:1-8, 2005
[3] Souza JP, Gülmezoglu AM, Vogel J, et al. Moving beyond [17] Roos JF, Doust J, Tett SE, Kirkpatrick CM. Diagnostic ac-
essential interventions for reduction of maternal mortality curacy of cystatin C compared to serum creatinine for the
(the WHO Multicountry Survey on Maternal and Newborn estimation of renal dysfunction in adults and children--a
Health): a cross-sectional study. Lancet 381:1747-1755, meta-analysis.  Clin Biochem 40:383-391, 2007
2013 [18] Dharnidharka VR, Kwon C, Stevens G. Serum cystatin C is
[4] Hansen AR, Barnés CM, Folkman J, McElrath TF. Maternal superior to serum creatinine as a marker of kidney func-
preeclampsia predicts the development of bronchopulmo- tion: a meta-analysis.  Am J Kidney Dis 40:221-226, 2002
nary dysplasia.  J Pediatr 156:532-536, 2010 [19] Zappitelli M, Greenberg JH, Coca SG, et al. Association
[5] Strand KM, Heimstad R, Iversen AC, et al. Mediators of the of definition of acute kidney injury by cystatin C rise with
association between pre-eclampsia and cerebral palsy: biomarkers and clinical outcomes in children undergoing
population based cohort study. BMJ 347:f4089, 2013 cardiac surgery.  JAMA Pediatr 169:583-591, 2015
[6] Prakash J, Ganiger VC, Prakash S, et al. Acute kidney injury [20] Yalcin S, Ulas T, Eren MA, et al. Relationship between oxi-
in pregnancy with special reference to pregnancy-specific dative stress parameters and cystatin C levels in patients
disorders: a hospital based study (2014-2016). J Nephrol with severe preeclampsia. Medicina (Kaunas ) 49:118-123,
31:79-85, 2018 2013
[7] Alge JL, Arthur JM. Biomarkers of AKI: a review of mecha- [21] Niraula A, Lamsal M, Baral N, et al. Cystatin-C as a marker
nistic relevance and potential therapeutic implications. Clin for renal impairment in preeclampsia.  J Biomark 2017:
J Am Soc Nephrol 10:147-155, 2015 7406959, 2017
[8] Vaidya VS, Ferguson MA, Bonventre JV. Biomarkers of acute [22] Abitbol CL, Seeherunvong W, Galarza MG, et al. Neonatal
kidney injury. Annu Rev Pharmacol Toxicol 48:463-493, kidney size and function in preterm infants: what is a true
2008 estimate of glomerular filtration rate? J Pediatr 164:1026-
[9] Jim B, Garovic VD. Acute kidney injury in pregnancy. Se- 1031, 2014
min Nephrol 37:378-385, 2017 [23] Gursoy AY, Tasci Y, Celik H, et al. The prognostic value of

344 www.krcp-ksn.org
 Wattanavaekin, et al. CysC, a biomarker of AKI, as a predictor of preterm labor

first-trimester cystatin C levels for gestational complica- culator.

tions.  J Perinat Med 44:295-299, 2016 [37] Cockcroft DW, Gault MH. Prediction of creatinine clear-
[24] Risch M, Purde MT, Baumann M, et al. High first-trimester ance from serum creatinine. Nephron 16:31-41, 1976
maternal blood cystatin C levels despite normal serum creat- [38] Mundt LA, Shanahan K. Chemical analysis of urine. In:
inine predict pre-eclampsia in singleton pregnancies. Scand Mundt LA, Shanahan K, eds. Graff’s Textbook of Routine
J Clin Lab Invest 77:634-643, 2017 Urinalysis and Body Fluids, 2nd edition. Philadelphia:
[25] Buderer NM. Statistical methodology: I. Incorporating the Wolters Kluwer/Lippincott Williams & Wilkins Health, 35-
prevalence of disease into the sample size calculation for 54, 2010
sensitivity and specificity.  Acad Emerg Med 3:895-900, 1996 [39] Kattah AG, Garovic VD. The management of hypertension
[26] Ngwenya S. Severe preeclampsia and eclampsia: incidence, in pregnancy.  Adv Chronic Kidney Dis 20:229-239, 2013
complications, and perinatal outcomes at a low-resource [40] National High Blood Pressure Education Program Working
setting, Mpilo Central Hospital, Bulawayo, Zimbabwe. Int J Group on High Blood Pressure in Pregnancy. Report of the
Womens Health 9:353-357, 2017 National High Blood Pressure Education Program Working
[27] Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet Group on high blood pressure in pregnancy. Am J Obstet
365:785-799, 2005 Gynecol 183:S1-S22, 2000
[28] Howson CP, Kinney MV, McDougall L, Lawn JE; Born Too [41] Padma Y, Aparna VB, Kalpana B, Ritika V, Sudhakar PR.
Soon Preterm Birth Action Group. Born too soon: preterm Renal markers in normal and hypertensive disorders of
birth matters.  Reprod Health 10 Suppl 1:S1, 2013 pregnancy in Indian women: a pilot study. Int J Reprod
[29] Lee AC, Kozuki N, Cousens S, et al. Estimates of burden Contracept Obstet Gynecol 2:514-520, 2013
and consequences of infants born small for gestational age [42] Strevens H, Wide-Swensson D, Grubb A. Serum cystatin C
in low and middle income countries with INTERGROWTH- is a better marker for preeclampsia than serum creatinine
st
21 standard: analysis of CHERG datasets. BMJ 358:j3677, or serum urate.  Scand J Clin Lab Invest 61:575-580, 2001
2017 [43] Franceschini N, Qiu C, Barrow DA, Williams MA. Cystatin
[30] Kiserud T, Benachi A, Hecher K, et al. The World Health C and preeclampsia: a case control study. Ren Fail 30:89-
Organization fetal growth charts: concept, findings, inter- 95, 2008
pretation, and application. Am J Obstet Gynecol 218:S619- [44] Babay Z, Al-Wakeel J, Addar M, et al. Serum cystatin C in
S629, 2018 pregnant women: reference values, reliable and superior
[31] Al-Turki HA, Abu-Heija AT, Al-Sibai MH. The outcome of diagnostic accuracy. Clin Exp Obstet Gynecol 32:175-179,
pregnancy in elderly primigravidas. Saudi Med J 24:1230- 2005
1233, 2003 [45] Novakov Mikic A, Cabarkapa V, Nikolic A, et al. Cystatin C
[32] Wall-Wieler E, Roos LL, Nickel NC. Adolescent pregnancy in pre-eclampsia. J Matern Fetal Neonatal Med 25:961-965,
outcomes among sisters and mothers: a population-based 2012
retrospective cohort study using linkable administrative [46] Bjornstad P, Cherney DZI. Kidney function can predict
data.  Public Health Rep 133:100-108, 2018 pregnancy outcomes. Clin J Am Soc Nephrol 12:1029-1031,
[33] Inker LA, Schmid CH, Tighiouart H, et al; CKD-EPI Inves- 2017
tigators. Estimating glomerular filtration rate from serum [47] Piccoli GB, Attini R, Vasario E, et al. Pregnancy and chronic
creatinine and cystatin C.  N Engl J Med 367:20-29, 2012 kidney disease: a challenge in all CKD stages.  Clin J Am Soc
[34] Levey AS, Stevens LA, Schmid CH, et al. A new equation Nephrol 5:844-855, 2010
to estimate glomerular filtration rate.  Ann Intern Med [48] Sturgiss SN, Wilkinson R, Davison JM. Renal reserve during
150:604-612, 2009 human pregnancy.  Am J Physiol 271(1 Pt 2):F16-F20, 1996
[35] Levey AS, Coresh J, Greene T, et al. Using standardized se- [49] Nevis IF, Reitsma A, Dominic A, et al. Pregnancy outcomes in
rum creatinine values in the modification of diet in renal women with chronic kidney disease: a systematic review. Clin
disease study equation for estimating glomerular filtration J Am Soc Nephrol 6:2587-2598, 2011
rate.  Ann Intern Med 145:247-254, 2006 [50] Piccoli GB, Cabiddu G, Attini R, et al. Risk of adverse preg-
[36] National Kidney Foundation, Inc. GFR Calculator. Available nancy outcomes in women with CKD. J Am Soc Nephrol
at: https://www.kidney.org/professionals/kdoqi/gfr_cal- 26:2011-2022, 2015

www.krcp-ksn.org 345
Kidney Res Clin Pract Vol. 37, No. 4, December 2018

[51] Knight EL, Verhave JC, Spiegelman D, et al. Factors influ- with standard measures and iohexol clearance. Diabetes
encing serum cystatin C levels other than renal function Care 25:2004-2009, 2002
and the impact on renal function measurement.  Kidney Int [56] Xia LH, Bing XG, An XT. Serum cystatin C assay for the de-
65:1416-1421, 2004 tection of early renal impairment in diabetic patients. J Clin
[52] Chu SC, Wang CP, Chang YH, et al. Increased cystatin C Lab Anal 18:31-35, 2004
serum concentrations in patients with hepatic diseases of [57] O’Riordan SE, Webb MC, Stowe HJ, et al. Cystatin C im-
various severities.  Clin Chim Acta 341:133-138, 2004 proves the detection of mild renal dysfunction in older pa-
[53] Fricker M, Wiesli P, Brändle M, Schwegler B, Schmid C. Im- tients.  Ann Clin Biochem 40:648-655, 2003
pact of thyroid dysfunction on serum cystatin C.  Kidney Int [58] Curhan G. Cystatin C: a marker of renal function or some-
63:1944-1947, 2003 thing more?  Clin Chem 51:293-294, 2005
[54] Wiesli P, Schwegler B, Spinas GA, Schmid C. Serum cystatin [59] Koenig W, Twardella D, Brenner H, Rothenbacher D. Plas-
C is sensitive to small changes in thyroid function. Clin ma concentrations of cystatin C in patients with coronary
Chim Acta 338:87-90, 2003 heart disease and risk for secondary cardiovascular events:
[55] Tan GD, Lewis AV, James TJ, et al. Clinical usefulness of more than simply a marker of glomerular filtration rate. Clin
cystatin C for the estimation of glomerular filtration rate Chem 51:321-327, 2005
in type 1 diabetes: reproducibility and accuracy compared

346 www.krcp-ksn.org