Objectives: Organ transplant is an effective treatment for patients with end-stage renal and hepatic failure. Increased use has introduced more emergency department admissions of infectious origin after transplant. Because infections usually manifest with simple complaints and fever, emergency physicians need laboratory tests and radiologic imaging procedures to quickly detect the presence and source of infection. Our aim was to analyze fever-related emergency admissions of renal and hepatic graft recipients and determine whether admitted patients had increased red blood cell distribution width and mean platelet volume levels.
Materials and Methods: We reviewed the medical records of renal and hepatic graft patients who presented to our emergency department with fever during a 4-year period. Our analyses included 150 patients in which complete blood count and C-reactive protein results were available and the source of infection was determined. We compared results with a control group of 150 transplant patients without any infectious findings.
Results: In the 150 solid-organ graft recipients who presented to our emergency department with fever, significant differences were observed versus control patients with respect to white blood cell count, neutrophil-to-lymphocyte ratio, red blood cell distribution width, mean platelet volume, and C-reactive protein levels (P < .05). We determined that C-reactive protein levels, red blood cell distribution width, mean platelet volume, and lymphocyte count were independent indicators of infection on multiple logistic regression analyses. We also determined that red blood cell distribution width had a specificity of 94% and a sensitivity of 26%.
Conclusions: We found a significantly higher red blood cell distribution width in emergency admissions of infectious origin of renal and hepatic graft recipients than in the control group (P < .001), suggesting that this measurement is a suitable marker of infection for the emergency setting by virtue of rapid availability of test results and lack of extra costs.
Key words : Fever, Transplant graft recipients’ infection, Emergencies
Organ transplant is a widely recognized and increasingly used effective treatment of end-stage renal and hepatic failure.1 The increased use of this treatment has resulted in more emergency department admissions of infectious origin after transplant. Because infections usually manifest with simple complaints and fever in transplant recipients,2 emergency department physicians need laboratory tests and radiologic imaging procedures to quickly detect the presence and source of infections in these patients. In the emergency department setting, these should be rapid and low-cost tests, enabling efficient use of resources, especially in countries with scarce resources.
Laboratory parameters such as leukocytosis and C-reactive protein (CRP) elevation are used by our emergency department and by many other institutions in our country for fever-related admissions of transplant recipients. Various markers including procalcitonin, interleukins 6 and 8, and lipopolysaccharide binding protein have not been yet introduced into daily clinical practice in emergency departments. Thus, there is a need for rapid, inexpensive markers for use worldwide. For example, red blood cell distribution width (RDW), mean platelet volume (MPV), and mean neutrophil-to-lymphocyte ratio are some parameters that can be readily calculated from a complete blood count. Their use as markers of sepsis also has been reported.3-5 The main advantage of all of these parameters is that they can be analyzed without the need for any additional costs or labor.
Herein, we aimed to analyze fever-related emergency department admissions of renal and hepatic graft recipients; we also aimed to determine whether there was an increase in RDW and MPV levels in these admissions.
Materials and Methods
We reviewed the medical records of renal and hepatic transplant patients who presented to the Department of Emergency Medicine at Baskent University with fever during a 4-year period from April 2011 to April 2015. Our study group included 150 patients who had complete blood cell count and CRP results and who had the source of fever or infection determined with laboratory tests, imaging modalities, and consultation notes of the relevant disease departments (eg, infectious diseases, respiratory diseases). We excluded patients with absent complete blood cell count and CRP measurements and those in which the source of fever could not be found. The control group consisted of 150 transplant patients without any infectious complaint or finding who presented to the Transplant Outpatient Clinic at Baskent University Ankara Hospital for routine purposes. After approval was obtained from the Baskent University Medical and Health Sciences Research Board, we retrospectively reviewed the medical records of all patients in the study and control groups using the hospital automation system.
We analyzed the study data with SPSS for Windows 17.0 software (SPSS: An IBM Company, version 17.0, IBM Corporation, Armonk, NY, USA). We presented the descriptive data as frequency (n), percentage (%), median, and interquartile range (the distance between the 25th and 75th percentiles). We used the Kolmogorov-Smirnov test to analyze the normality of the study data, and we observed that the study data were not normally distributed. Therefore, we used Mann-Whitney test for intergroup comparisons. We performed a multiple linear regression analysis to determine factors affecting infections. To determine the ability of RDW, MPV, and CRP to indicate an existing infection, we carried out receiver operating characteristic curve analysis, and we denoted values between 0.7 and 0.8 as acceptable, 0.8 and 0.9 as very good, and above 0.9 as excellent. P values < .05 were considered statistically significant.
Between April 2011 and April 2015, our emergency department received 150 solid-organ graft recipients with fever. Of these, 77 were women (51.3%) and 73 were men (48.7%). Seventy-three patients were hepatic graft recipients, and 77 were renal graft recipients. The control group consisted of 56 women (37.3%) and 93 men (62%). The median age of the patient and control groups were 40 (21) and 37 years (23). The most common type of infection was urinary tract infection (31.3%). Table 1 lists the infections detected in patients.
The laboratory parameters of the patients with infection and control groups are summarized in Table 2. We observed significant differences between both groups with respect to white blood cell count, neutrophil count, lymphocyte count, neutrophil-to-lymphocyte ratio, RDW, MPV, and CRP levels (P < .05).
We performed a multiple linear regression analysis to determine the independent factors indicating the existence of an infection (Table 3); we determined that CRP, RDW, MPV, and lymphocyte count were independent indicators of infection.
We additionally performed a receiver operating characteristic curve analysis to compare the abilities of CRP, RDW, and MPV to indicate infection, the latter of which we suggest can be used as a novel marker and which was significantly higher in the patients with infection than in the control group. The area under the curve for RDW was 0.762, 0.589 for MPV, and 0.956 for CRP; therefore, we considered the ability of CRP results to indicate infection as excellent and the ability of RDW to indicate infections as acceptable. We also determined that RDW had a specificity of 94% and a sensitivity of 26% for a threshold level of 18.25% that was just above the reference range (Figure 1).
Fever is regarded as the first symptom of a serious infection in transplant patients who are chronically immunosuppressed. It is therefore vital to determine the factors causing fever and start an appropriate treatment without delay in transplant patients presenting to emergency departments with fever. To accomplish this goal, extensive laboratory tests and radiologic imaging modalities are needed.2 One possible laboratory tests is CRP, a reportedly sensitive indicator of systemic inflammatory response.6 However, its level has been reported to be affected by several factors, including malignancy, sex, and age.7 Problems also arise because institutions that depend on social security payments cannot include CRP for reimbursement for patients needing this test in the emergency department setting. Thus, other low-cost biomarkers indicating infections in the emergency department are needed. In the present study designed to reach this goal, we found that RDW was significantly higher in solid-organ graft recipients than in the control group (P < .001). We suggest that RDW, which is a routinely measured and reported part of the complete blood count and thus without extra cost, can be used as a rapid biomarker to indicate infection in fever-related emergency department admissions. We believe that our results should be supported by future prospective, multicenter studies.
Mucsi and associates reported that elevated RDW levels predicted mortality in renal graft patients. The authors also stated that RDW may also be used for risk assessment in these patients.8 On the other hand, Balta and associates stressed that RDW cannot provide clear information regarding the inflammatory state of a patient without specific information from other markers such as CRP, neutrophil-to-lymphocyte ratio, or MPV.9 In this study, we measured CRP, neutrophil-to-lymphocyte ratio, and MPV in transplant patients and determined that all 3 markers were significantly higher than those in the control group (P < .05). We observed that RDW and MPV were independently correlated to an infection (P < .001). Based on this finding, we suggest that the evaluation of RDW in conjunction with neutrophil-to-lymphocyte ratio and MPV, both of which are readily measurable from complete blood cell counts and thus without extra cost or labor, would increase the reliability of the test.
It has been reported that a delay in collecting blood samples can lead to an erroneous RDW level.9 In our emergency department, blood samples are taken immediately after physical examination and sent to the biochemistry laboratory via a pneumatic system without delay. Complete blood cell count is completed within 45 seconds to 1 minute in a device dedicated for use for emergency cases. For this reason, no delay occurred in the RDW measurements that altered the results.
Thyroid dysfunction, nutritional deficiency, bone marrow dysfunction, and race are known to affect RDW levels.10 All of the patients in our study were of the same race; in addition, they had no thyroid dysfunction, nutritional deficiency, or bone marrow dysfunction.
Jo and associates showed that RDW predicted prognosis in sepsis. An elevated RDW level that occurs simultaneously with an inflammatory process is attributed to decreased red blood cell development by proinflammatory cytokines.3,11 We also believe that increased RDW levels resulted from this factor in our study.
In conclusion, we found a significantly higher RDW level in renal and hepatic graft recipients admitted to our emergency department with infection than in our control group (P < .001). We found an area under curve of 0.762 for RDW, indicating an acceptable ability of RDW to indicate infection. Based on the results of this study, we believe that RDW can be a useful marker of infection in renal and hepatic graft recipients admitted to an emergency department with fever. These results require confirmation in a future multicenter, prospective study. Our findings suggest that RDW is a suitable marker of infection in an emergency department by virtue of its rapid availability after complete blood cell count analysis and the lack of extra costs.
Volume : 15
Issue : 1
Pages : 61 - 64
DOI : 10.6002/ect.2015.0167
From the Departments of 1Emergency and 2General
Surgery, Faculty of Medicine, Baskent University, Ankara, Turkey
Acknowledgements: The authors have no conflicts of interest to disclose, and there was no funding for this study. This study was presented as poster presentation at the 11th Turkey Emergency Medicine Congress, 26-29 Oct 2015, Girne, North Cyprus.
Corresponding author: Afsin Emre Kayipmaz, Baskent University, Department of Emergency, Fevzi Cakmak Caddesi 10. Sokak, No: 45, Bahcelievler, Ankara 06490, Turkey
Phone: +90 312 203 6868
Fax: +90 312 223 7333
Table 1. Infections Shown in Transplant Recipients, as Detected by Laboratory Tests and Imaging Modalities
Table 2. Laboratory Parameters in Patient and Control Groups, Their Reference Ranges, and Statistical Analysis Results
Table 3. Multiple Linear Regression Analysis of the Correlations Among Mean Platelet Volume, Lymphocyte Count, Red Blood Cell Distribution Width, and C-Reactive Protein Levels
Figure 1. Receiver Operating Curve Analysis of C-Reactive Protein Level, Mean Platelet Volume, and Red Blood Cell Distribution Width