| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Brief Communications |
1 Department of Clinical Biochemistry and2 Division of Pediatric Cardiology, University of Bonn, Bonn, Germany;
aaddress correspondence to this author at: Institute for Clinical Biochemistry, University Hospital Bonn, Sigmund-Freud-Str. 25, D 53127 Bonn, Germany. Fax + 49 228 12159; e-mail berndt.zur{at}ukb.uni-bonn.de.
Abstract
Background: A 4-year-old boy and his father exhibited low oxygen saturation measured transcutaneously by pulse oximetry, a finding that could not be confirmed by arterial blood gas analysis. Both patients exhibited slight hemolysis in their blood, and the boy had a microcytic anemia. There was no evidence of hypoxemia or methemoglobinemia. Despite the normal results from the arterial blood gas analysis, a right-to-left-shunt was assumed in the boy until a cardiology examination excluded this diagnosis. Sleep apnea syndrome was suspected in the father and treated with nocturnal positive pressure respiration based on the low oxygen saturation values obtained with pulse oximetry. Only after consultation with our laboratory was a hemoglobin variant suspected and investigated.
Methods: We performed hemoglobin protein analysis by HPLC, electrophoretic separation, and spectrophotometry and DNA sequence analysis of the 
-globin gene.
Results: Both HPLC chromatographic separation and alkaline electrophoresis revealed a unique hemoglobin peak. In both patients, -globin gene sequencing revealed a mutation resulting in a histidine-to–aspartatic acid substitution at position 87. The low oxygen saturation measurement by pulse oximetry was due to hemoglobin Bonn oxyhemoglobin having an absorption peak at 668 nm, near the 660 nm measured by pulse oximeters.
Conclusion: Hemoglobin Bonn is a novel hemoglobin variant of the proximal -globin that results in falsely low oxygen saturation measurements with pulse oximetry.
A 4-year-old boy being treated for Morgagni hydatid had elective surgery under general anesthesia. Intraoperatively, a low O2 saturation value of 88% by pulse oximetry was noted, which increased to only 91% after insufflation with pure oxygen. Arterial blood gas analysis failed to confirm this low oxygen saturation. Nevertheless, a heart defect with right-to-left-shunt was suspected but was later excluded by duplex echocardiography at the Children’s Cardiology Department of the University of Bonn. The 41-year old father reported that during his own diagnostic examination for a possible sleep apnea syndrome discrepancies in oxygen saturation values between pulse oximetry and blood gas analysis were also found. In his case, despite the fact that oxygen saturation was normal by arterial blood gas analysis, the low oxygen saturation by pulse oximetry were used to assess the severity of his sleep apnea syndrome. Consequently, the father was treated with nocturnal positive pressure respiration. In the autumn of 2006, the boy presented for follow-up at the Children’s Cardiology Department of the University of Bonn.
Neither father nor son had any clinical indications of hypoxemia or reduced fitness. Pulse oximetry [NELLCOR OxiMax N-550 (Tyco)] via a clip on the index finger revealed oxygen saturation measurement by pulse oximetry to be 88% in the father and 87% in the son. The pulse oximetry device uses light-emitting diodes at 660 nm and 940 nm. Venous blood samples from father and son were analyzed using the XE 2100 (Sysmex), the Dimension RXL (Dade Behring), the BN II Nephelometer (Dade Behring), and the Vitros 250 Analyzer (Ortho-Clinical Diagnostics) for hemogram, chemistry, proteins, and bilirubin analyses, respectively. Arterial blood was analyzed with the ABL 735 blood analyzer (Radiometer). Free hemoglobin was measured with the spectrophotometer LS 500 (Lange). The son had a hemoglobin concentration of 108 g/L (reference interval 111–144 g/L), leading to a diagnosis of mildly hypochromic microcytic anemia with a normal ferritin value of 25 ug/L (15–142 ug/L). Concentrations of nonconjugated bilirubin (4.6 mg/L) (<2.5 mg/L) as well as of free hemoglobin (115 mg/L) (<50 mg/L) were slightly increased, whereas haptoglobin (0.28 g/L) was marginally below the reference interval (0.3–2.0 g/L). Reticulocytes were increased at 3.3% (0.5%–2%). In the father, slight increases in nonconjugated bilirubin (4.9 mg/L) and in free hemoglobin (119 mg/L) as well as a mild reticulocytosis (2.2%) was observed. Glucose-6-phosphate dehydrogenase activity was normal. In both patients, the methemoglobin fraction was only 0.1%. Arterial blood gas analysis revealed O2 saturation of 95.8%, (pO2 86.2 mm Hg) and 97.3%, (pO2 99.2 mm Hg) in the father and son, respectively. Physical examination, electrocardiogram, echocardiograph, and chest x-ray revealed no clinical manifestations in the father or son. Normal oxygen saturation was observed by pulse oxymetry in the father’s brother and the son’s sister. The laboratory was contacted, leading to identification of a hemoglobin anomaly in both father and son.
In both patients, chromatographic separation of hemoglobin using the Variant II HPLC (Bio-Rad) showed a distinct peak at 0.67–0.70 min, immediately before the HbA1c peak.
Gel electrophoresis was performed with Hydrasys (Sebia) using a commercially available method (Hydragel Hemoglobin). Alkaline electrophoresis revealed an unknown band cathodally before the HbA0 peak.
Genomic DNA was isolated by the use of the QIAmp DNA Blood Kit (Qiagen). PCR amplification of the respective products and bidirectional automated sequence analysis was performed as recently described (1). PCR primers were designed to selectively amplify the human 1- and -globin genes. The forward primer (1/2-F, 5'-CCAAGCATAAACCCTGGCGC-3') was complementary to a conserved region of the promoter and exon 1 of the genes, whereas the reverse primers were complementary to a nonhomologous part of the 3' untranslated region (1-R, 5'-CACGGGGGTACGGGTGCAG-3'; 2-R, 5'-AGGAAGGGCCGGTGCAAGG-3'). PCR primers also served as sequencing primers, and exon 2 of the amplified -gene products was analyzed with 2 universal primers (1/2-2F, 5'-CACAGGCCACCCTCAACCGT-3'; 1/2–2R, 5'-GCCATCTCGCCCCTCGACC-3'). In father and son, sequencing of the -globin gene revealed a point mutation c.C299G (Genbank acc. no. NM_000558) in exon 2 of the hemoglobin, alpha 1 (HBA1) gene, resulting in an H87D exchange. Spectrophotometric measurement of the pretreated capillary blood using the LS 500 spectrophotometer (Lange) revealed an additional absorption maximum of oxyhemoglobin at 668 nm, while the absorption curve of deoxyhemoglobin remained inconspicuous (Fig. 1
).
|
Pulse oximetry is commonly used for rapid measurement of pulse rate and oxygen saturation. Because of their absorption spectra, however, hemoglobin variants and some derivatives of normal hemoglobin may interfere with pulse oximeter measurement (2)(5). For example, hemoglobins Köln and Cheverly also produce falsely-low oxygen saturation values by pulse oximetry (6)(7)(8).
In commonly applied pulse oximeters, absorption is measured at 660 m and 940 nm, where the largest differences in absorption of oxyhemoglobin and deoxyhemoglobin can be found. Hemoglobin Bonn has an additional oxyhemoglobin absorption maximum at 668 nm, resulting in a falsely-low estimation of oxygen saturation (Fig. 1
) (4)(9).
Despite normal oxygen saturation findings from the blood gas analyzer, diagnoses were made in both father and son that resulted in relatively complex and cost-intensive workups or therapy as well as considerable psychological stress. The father eventually received therapeutically incorrect treatment (nocturnal positive pressure respiration for suspected sleep apnea syndrome). Only upon contacting the laboratory was a hemoglobin anomaly considered.
Hemoglobin Bonn can be easily identified by chromatography, electrophoresis, or spectrophotometry. With the identification of this and other hemoglobins that interfere with pulse oximeter measurement, complex and expensive examinations might be avoided.
Acknowledgments
Grant/funding Support: None declared.
Financial Disclosures: None declared.
Acknowledgment: For excellent technical assistance our thanks go to J. Koksch and Mrs. M. Schmidt.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
