The Laryngoscope C 2014 The American Laryngological, V
Rhinological and Otological Society, Inc.
False Negative b-2 Transferrin in the Diagnosis of Cerebrospinal Fluid Leak in the Presence of Streptococcus pneumoniae Maya Korem, MD; Haim Ovadia, PhD; Iddo Paldor, MD; Allon E. Moses, MD; Colin Block, MD; Ron Eliashar, MD; Nir Hirshoren, MD Objectives/Hypothesis: The objectives of this study were to examine the presence of b-2 transferrin (b2TRNSF) in cerebrospinal fluid (CSF) contaminated in vitro by various bacteria and explore the mechanism (passive or active) responsible for b2TRNSF elimination. Early diagnosis of CSF leakage may change treatment decisions and minimize the risk of meningitis and encephalitis. b2TRNSF is a protein present exclusively in CSF. Its detection is highly useful in cases of CSF leakage, although it has never been examined in the presence of central nervous system infection. Study Design: Prospective patient analysis. Methods: Sterile CSF drawn from patients was contaminated in vitro with several microorganisms chosen for their ability to cause neurosurgical-related infections: Streptococcus pneumoniae, methicillin-sensitive Staphylococcus aureus, Staphylococcus epidermidis, and Pseudomonas aeruginosa. b2TRNSF was examined at two time points: following immediate inoculation (t0) and following an overnight incubation (t18) over various bacterial concentrations. Samples of CSF inoculated with S pneumoniae were also examined in the presence of ciprofloxacin. For b2TRNSF analysis we used immunoblotting electrophoresis and enzyme-linked immunosorbent assay (ELISA). Results: CSF samples collected from nine patients were analyzed. b2TRNSF was not detected following S pneumoniae inoculation at both time points when immunoblotting electrophoresis was used. Quantitative analysis using ELISA demonstrated significant b2TRNSF concentration decrease. The addition of ciprofloxacin led to the same results. Conclusions: CSF leak detection using b2TRNSF may be deceiving in the presence of a S pneumoniae cerebral nervous system infection. A passive process is suggested, as b2TRNSF disappeared either immediately or following incubation with inactive bacteria. Key Words: b-2 transferrin, cerebrospinal fluid leakage, electrophoresis, enzyme-linked immunosorbent assay, Streptococcus pneumoniae. Level of Evidence: NA Laryngoscope, 125:556–560, 2015
INTRODUCTION Cerebrospinal fluid (CSF) leakage secondary to fistula formation between the dura matter and skull base may result from traumatic head injury, iatrogenic injury, or spontaneously (associated with low or high intracranial pressure, as in congenital anomalies, brain tumors, cysts, and hydrocephalus).1–4 Free passage of bacterial flora from the nasal cavity and paranasal sinuses through a CSF fistula into the cranium may pose a risk for meningitis and encephalitis. Meningitis may super-
From the Department of Clinical Microbiology and Infectious Diseases (M.K., A.E.M., C.B.), Department of Neurology (H.O.), the Agnes Ginges Center for Human Neurogenetics, Department of Neurosurgery (I.P.), and the Department of Otolaryngology/Head & Neck Surgery (R.E., N.H.), Hebrew University School of Medicine–Hadassah Medical Center, Jerusalem, Israel. Editor’s Note: This Manuscript was accepted for publication August 29, 2014. This study was supported by the Hebrew University School of Medicine–Hadassah Medical Center Grant for Young Clinician (N.H.). The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Nir Hirshoren, MD, Department of Otolaryngology/Head & Neck Surgery, Hadassah Ein-Kerem, Jerusalem, 91120, Israel. E-mail: [email protected] DOI: 10.1002/lary.24940
Laryngoscope 125: March 2015
556
vene in as many as 19% of such cases.5 Thus, early diagnosis and treatment of CSF leakage may minimize the risk of a lethal infection. The b-2 transferrin (b2TRNSF) protein was introduced as a marker for CSF by Meurman et al. in 1979.6 b2TRNSF detection is a highly reliable sensitive diagnostic method for detecting CSF leakage.7,8 This glycoprotein is a desialylated isoform of transferrin without neuraminic acid and is found almost exclusively in the CSF. It is believed that b2TRNSF is converted from the b-1 isoform via cerebral neuraminidase rather than produced de novo.9 Therefore, the presence of b2TRNSF may be used as a marker for CSF leakage even when the CSF is mixed with blood or with other secretions.1,10–12 A common laboratory technique for b2TRNSF detection is immunofixation electrophoresis. The presence of two transferrin bands in b-globulin fraction indicates a positive result for CSF. To achieve good sensitivity and compensate for a low CSF b2TRNSF concentration, 2 to 5 mL are required,13 whereas only 2 lL10 are required if immunoblotting electrophoresis is used. The ability to detect b2TRNSF in CSF samples of patients suffering from bacterial central nervous system (CNS) infections has never been explored. We had two patients with pneumococcal meningitis and clinical CSF Korem et al.: Misdiagnosis of Cerebrospinal Fluid Leak
leakage following brain surgery, in whom b2TRNSF in the nasal fluid was not detectable using immunoblotting electrophoresis. We suspect that in selected cases of CNS bacterial infections, b2TRNSF might be adsorbed, degraded, or consumed by bacteria; therefore, its assessment in a bacterial milieu could be misleading. The goal of this study was to examine the presence of b2TRNSF in CSF inoculated in vitro by various bacteria.
TABLE I. Characteristics of Nine Patients Prospectively Recruited for the Study. Patient Name (Initials)
Gender
EE
M
74
NPH evaluation
YY
M
72
NPH evaluation
CG
M
37
PL
F
36
AP
M
67
BM
F
67
Postoperative (arachnoid cyst surgery) CSF leak Postoperative (pituitary surgery) CSF leak Postoperative (pineal surgery) CSF leak NPH evaluation
AA
F
79
NPH evaluation
KN BS
F F
80 83
NPH evaluation NPH evaluation
Age, yr
MATERIALS AND METHODS Patients Between 2012 and 2013, patients with CSF continuous drainage placement either for evaluation or treatment of different medical conditions were prospectively recruited to participate in the study. The study was approved and conducted in accordance with the guidelines of the local institutional review board (Helsinki Committee). All patients agreed and signed an informed consent form.
Cerebrospinal Fluid Waste CSF from neurosurgical patients with a CSF drainage device was drawn and brought immediately to the microbiology laboratory, where it was stored at 4 C for no longer than 48 hours. Contaminated or infected CSF was excluded from the study by plating each sample overnight on blood agar.
Bacterial Preparation We used several bacterial isolates obtained from patients admitted to the hospital during the study period that were chosen for their ability to cause meningeal neurosurgical-related infection: Streptococcus pneumoniae, methicillin-sensitive Staphylococcus aureus, Staphylococcus epidermidis, and Pseudomonas aeruginosa. Cultures were prepared by inoculating a single colony from a blood agar plate (Novamed, Jerusalem, Israel), into 6 mL of tryptic soy broth (TSB) (Novamed) and incubating overnight (18 hours) at a temperature of 37 C. The cultures were added to at least 4 mL of sterile CSF until an optical density (OD) of 1 to 3.5 McFarland standard was obtained. The samples were subjected to centrifugation (4,000 rpm, 15 minutes) both immediately or following an overnight (18 hours) incubation at 37 C, allowing the presence of b2TRNSF to be examined at two time points defined as1 immediately upon CSF inoculation (t0)2 and extended CSF incubation (t18), over at least two different ODs for each sample. One milliliter of each supernatant was kept at 220 C until b2TRNSF was examined. To explore whether a passive or active mechanism is responsible for b2TRNSF elimination, a sample of CSF inoculated with S pneumoniae was also evaluated as described in the presence of 0.75 mg/L ciprofloxacin immediately and following 18 hours incubation at 37 C. The minimal inhibitory concentration of ciprofloxacin for the examined bacteria was 0.75 lg/mL.
Controls TSB, sterile CSF, and sterile CSF with TSB at different dilutions identical to those used with the bacteria served as controls. The samples were processed and examined as described above. Sterility of the samples was assured by plating the samples on a blood agar overnight at 37 C and examination for bacterial growth. Bacterial counts were performed for all samples using serial 10-fold dilutions in normal saline as described before.14
Laryngoscope 125: March 2015
Indication for Continued Drainage Insertion
All patients had continuous drainage placement either for evaluation or treatment of different medical conditions. CSF 5 cerebrospinal fluid; F 5 female; M 5 male; NPH 5 normal pressure hydrocephalus.
b2TRNSF Measurements b2TRNSF presence was determined by immunoblotting electrophoresis performed on agarose gels with immunofixation using a specific rabbit anti-human b2TRNSF (Hydrasys, Sebia, France). Sterile CSF samples and TSB samples were defined as positive and negative controls, respectively. If b-1 transferrin and b2TRNSF were detected in drainage fluids, the specimen was presumed to be CSF. ELISA was conducted in all CSF samples where electrophoresis had failed to detect b2TRNSF. b2TRNSF concentrations were determined using human b2TRNSF enzyme-linked immunosorbent assay (ELISA) kit (MyBioSource, San Diego, CA). This assay employs the quantitative sandwich enzyme immunoassay technique with a detection range of 1.25 mg/mL to 80 mg/mL. Color develops in proportion to the amount of b2TRNSF, and the intensity of the color is measured.
Statistical Analysis Data were analyzed by unpaired t test, and P values
Rhinological and Otological Society, Inc.
False Negative b-2 Transferrin in the Diagnosis of Cerebrospinal Fluid Leak in the Presence of Streptococcus pneumoniae Maya Korem, MD; Haim Ovadia, PhD; Iddo Paldor, MD; Allon E. Moses, MD; Colin Block, MD; Ron Eliashar, MD; Nir Hirshoren, MD Objectives/Hypothesis: The objectives of this study were to examine the presence of b-2 transferrin (b2TRNSF) in cerebrospinal fluid (CSF) contaminated in vitro by various bacteria and explore the mechanism (passive or active) responsible for b2TRNSF elimination. Early diagnosis of CSF leakage may change treatment decisions and minimize the risk of meningitis and encephalitis. b2TRNSF is a protein present exclusively in CSF. Its detection is highly useful in cases of CSF leakage, although it has never been examined in the presence of central nervous system infection. Study Design: Prospective patient analysis. Methods: Sterile CSF drawn from patients was contaminated in vitro with several microorganisms chosen for their ability to cause neurosurgical-related infections: Streptococcus pneumoniae, methicillin-sensitive Staphylococcus aureus, Staphylococcus epidermidis, and Pseudomonas aeruginosa. b2TRNSF was examined at two time points: following immediate inoculation (t0) and following an overnight incubation (t18) over various bacterial concentrations. Samples of CSF inoculated with S pneumoniae were also examined in the presence of ciprofloxacin. For b2TRNSF analysis we used immunoblotting electrophoresis and enzyme-linked immunosorbent assay (ELISA). Results: CSF samples collected from nine patients were analyzed. b2TRNSF was not detected following S pneumoniae inoculation at both time points when immunoblotting electrophoresis was used. Quantitative analysis using ELISA demonstrated significant b2TRNSF concentration decrease. The addition of ciprofloxacin led to the same results. Conclusions: CSF leak detection using b2TRNSF may be deceiving in the presence of a S pneumoniae cerebral nervous system infection. A passive process is suggested, as b2TRNSF disappeared either immediately or following incubation with inactive bacteria. Key Words: b-2 transferrin, cerebrospinal fluid leakage, electrophoresis, enzyme-linked immunosorbent assay, Streptococcus pneumoniae. Level of Evidence: NA Laryngoscope, 125:556–560, 2015
INTRODUCTION Cerebrospinal fluid (CSF) leakage secondary to fistula formation between the dura matter and skull base may result from traumatic head injury, iatrogenic injury, or spontaneously (associated with low or high intracranial pressure, as in congenital anomalies, brain tumors, cysts, and hydrocephalus).1–4 Free passage of bacterial flora from the nasal cavity and paranasal sinuses through a CSF fistula into the cranium may pose a risk for meningitis and encephalitis. Meningitis may super-
From the Department of Clinical Microbiology and Infectious Diseases (M.K., A.E.M., C.B.), Department of Neurology (H.O.), the Agnes Ginges Center for Human Neurogenetics, Department of Neurosurgery (I.P.), and the Department of Otolaryngology/Head & Neck Surgery (R.E., N.H.), Hebrew University School of Medicine–Hadassah Medical Center, Jerusalem, Israel. Editor’s Note: This Manuscript was accepted for publication August 29, 2014. This study was supported by the Hebrew University School of Medicine–Hadassah Medical Center Grant for Young Clinician (N.H.). The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Nir Hirshoren, MD, Department of Otolaryngology/Head & Neck Surgery, Hadassah Ein-Kerem, Jerusalem, 91120, Israel. E-mail: [email protected] DOI: 10.1002/lary.24940
Laryngoscope 125: March 2015
556
vene in as many as 19% of such cases.5 Thus, early diagnosis and treatment of CSF leakage may minimize the risk of a lethal infection. The b-2 transferrin (b2TRNSF) protein was introduced as a marker for CSF by Meurman et al. in 1979.6 b2TRNSF detection is a highly reliable sensitive diagnostic method for detecting CSF leakage.7,8 This glycoprotein is a desialylated isoform of transferrin without neuraminic acid and is found almost exclusively in the CSF. It is believed that b2TRNSF is converted from the b-1 isoform via cerebral neuraminidase rather than produced de novo.9 Therefore, the presence of b2TRNSF may be used as a marker for CSF leakage even when the CSF is mixed with blood or with other secretions.1,10–12 A common laboratory technique for b2TRNSF detection is immunofixation electrophoresis. The presence of two transferrin bands in b-globulin fraction indicates a positive result for CSF. To achieve good sensitivity and compensate for a low CSF b2TRNSF concentration, 2 to 5 mL are required,13 whereas only 2 lL10 are required if immunoblotting electrophoresis is used. The ability to detect b2TRNSF in CSF samples of patients suffering from bacterial central nervous system (CNS) infections has never been explored. We had two patients with pneumococcal meningitis and clinical CSF Korem et al.: Misdiagnosis of Cerebrospinal Fluid Leak
leakage following brain surgery, in whom b2TRNSF in the nasal fluid was not detectable using immunoblotting electrophoresis. We suspect that in selected cases of CNS bacterial infections, b2TRNSF might be adsorbed, degraded, or consumed by bacteria; therefore, its assessment in a bacterial milieu could be misleading. The goal of this study was to examine the presence of b2TRNSF in CSF inoculated in vitro by various bacteria.
TABLE I. Characteristics of Nine Patients Prospectively Recruited for the Study. Patient Name (Initials)
Gender
EE
M
74
NPH evaluation
YY
M
72
NPH evaluation
CG
M
37
PL
F
36
AP
M
67
BM
F
67
Postoperative (arachnoid cyst surgery) CSF leak Postoperative (pituitary surgery) CSF leak Postoperative (pineal surgery) CSF leak NPH evaluation
AA
F
79
NPH evaluation
KN BS
F F
80 83
NPH evaluation NPH evaluation
Age, yr
MATERIALS AND METHODS Patients Between 2012 and 2013, patients with CSF continuous drainage placement either for evaluation or treatment of different medical conditions were prospectively recruited to participate in the study. The study was approved and conducted in accordance with the guidelines of the local institutional review board (Helsinki Committee). All patients agreed and signed an informed consent form.
Cerebrospinal Fluid Waste CSF from neurosurgical patients with a CSF drainage device was drawn and brought immediately to the microbiology laboratory, where it was stored at 4 C for no longer than 48 hours. Contaminated or infected CSF was excluded from the study by plating each sample overnight on blood agar.
Bacterial Preparation We used several bacterial isolates obtained from patients admitted to the hospital during the study period that were chosen for their ability to cause meningeal neurosurgical-related infection: Streptococcus pneumoniae, methicillin-sensitive Staphylococcus aureus, Staphylococcus epidermidis, and Pseudomonas aeruginosa. Cultures were prepared by inoculating a single colony from a blood agar plate (Novamed, Jerusalem, Israel), into 6 mL of tryptic soy broth (TSB) (Novamed) and incubating overnight (18 hours) at a temperature of 37 C. The cultures were added to at least 4 mL of sterile CSF until an optical density (OD) of 1 to 3.5 McFarland standard was obtained. The samples were subjected to centrifugation (4,000 rpm, 15 minutes) both immediately or following an overnight (18 hours) incubation at 37 C, allowing the presence of b2TRNSF to be examined at two time points defined as1 immediately upon CSF inoculation (t0)2 and extended CSF incubation (t18), over at least two different ODs for each sample. One milliliter of each supernatant was kept at 220 C until b2TRNSF was examined. To explore whether a passive or active mechanism is responsible for b2TRNSF elimination, a sample of CSF inoculated with S pneumoniae was also evaluated as described in the presence of 0.75 mg/L ciprofloxacin immediately and following 18 hours incubation at 37 C. The minimal inhibitory concentration of ciprofloxacin for the examined bacteria was 0.75 lg/mL.
Controls TSB, sterile CSF, and sterile CSF with TSB at different dilutions identical to those used with the bacteria served as controls. The samples were processed and examined as described above. Sterility of the samples was assured by plating the samples on a blood agar overnight at 37 C and examination for bacterial growth. Bacterial counts were performed for all samples using serial 10-fold dilutions in normal saline as described before.14
Laryngoscope 125: March 2015
Indication for Continued Drainage Insertion
All patients had continuous drainage placement either for evaluation or treatment of different medical conditions. CSF 5 cerebrospinal fluid; F 5 female; M 5 male; NPH 5 normal pressure hydrocephalus.
b2TRNSF Measurements b2TRNSF presence was determined by immunoblotting electrophoresis performed on agarose gels with immunofixation using a specific rabbit anti-human b2TRNSF (Hydrasys, Sebia, France). Sterile CSF samples and TSB samples were defined as positive and negative controls, respectively. If b-1 transferrin and b2TRNSF were detected in drainage fluids, the specimen was presumed to be CSF. ELISA was conducted in all CSF samples where electrophoresis had failed to detect b2TRNSF. b2TRNSF concentrations were determined using human b2TRNSF enzyme-linked immunosorbent assay (ELISA) kit (MyBioSource, San Diego, CA). This assay employs the quantitative sandwich enzyme immunoassay technique with a detection range of 1.25 mg/mL to 80 mg/mL. Color develops in proportion to the amount of b2TRNSF, and the intensity of the color is measured.
Statistical Analysis Data were analyzed by unpaired t test, and P values
