The impacts of thyroid function on the diagnostic accuracy of Cystatin C to detect acute kidney injury in ICU patients: a prospective, observational study

Wang et al. Critical Care 2014, 18:R9 http://ccforum.com/content/18/1/R9 RESEARCH Open Access The impacts of thyroid function on the diagnostic accuracy of Cystatin C to detect acute kidney injury in ICU patients: a prospective, observational study Feilong Wang1†, Wenzhi Pan2†, Hairong Wang1†, Yu Zhou1, Shuyun Wang1 and Shuming Pan1* Abstract Introduction: Cystatin C (Cysc) could be affected by thyroid function both in vivo and in vitro and thereby may have limited ability to reflect renal function. We aimed to assess the association between Cysc and thyroid hormones as well as the effect of thyroid function on the diagnostic accuracy of Cysc to detect acute kidney injury (AKI). Methods: A total of 446 consecutive intensive care unit (ICU) patients were screened for eligibility in this prospective AKI observational study. Serum Cysc, thyroid hormones and serum creatinine (Scr) were measured upon entry to the ICU. We also collected each patient's baseline characteristics including the Acute Physiology and Chronic Health Evaluation II (APACHE-II) score. The diagnostic performance of Cysc was assessed from the area under the receiver operator characteristic curve (AUC) in each quartile of thyroid hormone(s). Results: A total of 114 (25.6%) patients had a clinical diagnosis of AKI upon entry to the ICU. The range of free thyroxine (FT4) value was 4.77 to 39.57 pmol/L. Multivariate linear regression showed that age (standardized beta = 0.128, P < 0.0001), baseline Scr level (standardized beta = 0.290, P < 0.0001), current Scr (standardized beta = 0.453, P < 0.0001), albumin (standardized beta = −0.086, P = 0.006), and FT4 (standardized beta = 0.062, P = 0.039) were related with Cysc. Patients were divided into four quartiles based on FT4 levels. The AUC for Cysc in detecting AKI in each quartile were as follows: 0.712 in quartile I, 0.754 in quartile II, 0.829 in quartile III and 0.797 in quartile IV. There was no significant difference in the AUC between any two groups (all P > 0.05). The optimal cut-off value of Cysc for diagnosing AKI increased across FT4 quartiles (1.15 mg/L in quartile I, 1.15 mg/L in quartile II, 1.35 mg/L in quartile III and 1.45 mg/L in quartile IV). Conclusions: There was no significant impact of thyroid function on the diagnostic accuracy of Cysc to detect AKI in ICU patients. However, the optimal cut-off value of Cysc to detect AKI could be affected by thyroid function. Introduction Acute kidney injury (AKI) is a prevalent problem and still a big challenge to both the developed and developing world [1]. About two-thirds of intensive care unit (ICU) patients develop an episode of AKI during their ICU stay [2]. Both short-term and long-term mortality were higher in ICU patients with AKI than those without [3-5]. Studies have found that early detection and treatment of AKI may improve outcomes [2]. Thus, * Correspondence: † Equal contributors 1 Department of Emergency, Xinhua Hospital Affiliated to Shanghai Jiaotong University School of Medcine, NO.1665, Kongjiang Road, Shanghai 200092, China Full list of author information is available at the end of the article timely diagnosis of AKI development after renal insult is urgent. Cystatin C (Cysc), a 13-kDa cysteine proteinase inhibitor, is freely filtered at the glomerulus and neither secreted nor reabsorbed by renal tubules. This physiological feature makes Cysc an ideal glomerular filtration biomarker. During the past few years, some studies have focused on the potential value of Cysc for the diagnosis and early detection of AKI [6-14]. However, these studies have reported conflicting results. Some studies reported good discrimination for Cysc in the early detection of AKI in various patient populations [7-9,14,15], while other studies found that Cysc had poor or moderate ability to predict AKI [6,10-13]. Besides that, there was no consensus about the appropriate cut-off value for using Cysc to diagnose or predict AKI [6-14,16]. These inconsistent © 2014 Wang et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Wang et al. Critical Care 2014, 18:R9 http://ccforum.com/content/18/1/R9 results limit the usefulness of Cysc in the early detection of AKI in clinical practice. Cysc is produced by all nucleated cells in the human body at a relatively constant rate [17]. However, recent studies found that thyroid hormones could stimulate the production of Cysc in vitro [18,19]. Moreover, clinical studies also found that Cysc was significantly associated with thyroid function [20-24]. The impact of thyroid hormones on the diagnostic value of Cysc in detecting AKI has raised concerns in clinical practice [25-27]. To the best of our knowledge, whether thyroid hormones are related to the level of Cysc in ICU patients has never been investigated. The effect of thyroid hormones on the diagnostic accuracy and threshold of Cysc in predicting AKI has also not been defined. Therefore, we undertook a prospective, observational study in a large population of unselected ICU patients to assess: 1) the relationship between Cysc and thyroid hormones; and 2) the effect of thyroid function on the diagnostic value of Cysc in detecting AKI. Material and methods Participants This prospective study recruited consecutive patients 18 years old and older hospitalized in the ICU of Xinhua Hospital affiliated with Shanghai Jiaotong University School of Medicine between April 2011 and May 2012, including medical and trauma patients. We decided a priori to exclude patients according to the following criteria: 1) past history of any thyroid diseases, such as hyperthyroidism, hypothyroidism and thyroid tumors; 2) thyroid nodule found by physical examination when admitted to ICU; 3) pregnancy within the previous six months; 4) undergoing any hormone replacement therapy except insulin use; 5) pre-existing severe renal disease (serum creatinine (Scr) >300 umol/L) or pre-existing dialysis; and 6) undergoing continuous renal replacement therapy (CRRT) in the four weeks before the blood sample was collected. Patients who died or were discharged from the ICU within four hours of admission were also excluded because data collection was difficult for these patients. The Shanghai Jiaotong University Xinhua Hospital Ethics Committee approved the study and waived the requirement for informed consent, because this was an observational study and all laboratory indices observed were commonly measured for all patients in our ICU department. Definition of acute kidney injury The patients were diagnosed as having AKI by using the stage 1 AKI criteria of the Acute Kidney Injury Network (AKIN) classification: new-onset of at least 1.5-fold increase or ≥0.3 mg/dL (26.5 umol/L) increment of SC (...truncated)