Prof. Ray Iles Presenting the mysterious case of “Bad Blood”

Prof. Ray Iles of MAP Sciences is presenting at Cambridge University & Addenbrookes Hospital Grand Round with Consultant Haematologist Dr Martin Besser.

“Bad Blood” is the case report of a lady poisoned by her own self-medication, resulting in Sulphur Haemoglobin in her blood which was only possible to diagnose using MAP Sciences MALDI ToF mass spectrometry technology.


Sulphated haemoglobin (SulfHb) is a rare entity caused by irreversible sulphation of the haem moiety in haemoglobin, which leads to a similar clinical presentation to methylated haemoglobin (MetHb). The diagnosis of SulfHb is challenging; it is often difficult to distinguish from MetHb in the arterial blood gas analysers in common use in the UK due to overlap in the absorption spectra of the two species. The presence of a SulfHb was suspected in a 73-year-old lady with low oxygen saturation (SaO2 ~75%), central cyanosis and normal arterial oxygen partial pressure (pO2 ~12kPa). Repeated arterial blood gas analysis on different systems returned error messages for MetHb quantification. There was improvement in oxygen saturation and cyanosis after exchange transfusion when the patient represented with atrial fibrillation and dyspnoea. Spectrophotometry of the patient’s blood was suggestive of the presence of SulfHb, with peak absorption at 620nm. Matrix-assisted laser desorption ionisation time-of-flight mass spectrometry (MALDI-ToF MS) was undertaken on whole blood samples from the patient pre- and post-transfusion, alongside normal controls. These demonstrated the presence of a SulfHb in the patient’s blood, specifically identifying sulphur, sulphur monoxide and sulphur dioxide bound to the haem moiety; levels fell after exchange transfusion. MALDI-ToF MS represents a new, rapid, sensitive and specific diagnostic test for the rare haematological syndrome of SulfHb. In addition, it can identify different sulphur compounds bound to haem, providing useful diagnostic clues as to the source of SulfHb where this is unclear.


A 73-year-old lady presented with a short history of cyanosis. She was found to have peripheral oxygen saturations of around 75% and cyanosis of her lips, tongue and fingers, with otherwise normal examination findings. The patient underwent extensive investigation by the local Cardiology and Respiratory services, with ultimate exclusion of primary cardiac or respiratory pathology. Her oxygen saturation remained low on repeated arterial blood gas sampling, despite normal PaO2 levels (around 12kPa). Methylated haemoglobin was suspected, but repeated error messages for MetHb quantification were returned by arterial blood gas machines made by different manufacturers in three separate hospitals. Haemoglobin high performance liquid chromatography (HPLC) and capillary electrophoresis results were normal. Analysis of the patient’s haemoglobin by a National Reference Laboratory returned normal results, including sequencing of her globin genes. Screening for paroxysmal nocturnal haemoglobinuria and glucose-6-phosphate dehydrogenase deficiency were negative. After a subsequent presentation with fast atrial fibrillation with dyspnoea, the patient underwent an exchange red cell transfusion, with improvement in her symptoms, cyanosis and oxygen saturation. The blood venesected from the patient was noted to be dark brown in colour (Figure 1). A literature review identified SulfHb as another potential cause of the patient’s clinical signs. She gave a history of rheumatoid arthritis, migraines and severe constipation, and potential sources of sulphur identified among her medications were the migraine drug rizatriptan, and a compound containing Epsom salts (magnesium sulphate) that was prescribed for constipation. A clinical diagnostic biotechnology startup company, MAP Sciences, was approached to undertake analysis of the patient’s blood to establish if SulfHb was detectable.



Visible light absorption spectral analysis of patient whole blood
20µl of whole blood samples were lysed in 2ml of ddH2O to causes cell lysis but also release of haem moieties from the globin proteins; at a dilution of 1/100 the main visible colour is due to the abundant haem moieties. Visible spectral absorption was measured at 10nm intervals between 500 and 700nm, blanking at each wavelength against ddH2O. Spectra absorption was normalised against peak absorption at 530nm to correct for variation in total Hb concentration of individual samples.

MALDI-ToF MS of blood for haem adducts
Control blood samples, along with the patient’s blood samples at presentation, immediately prior to exchange transfusion, and post-exchange transfusion were all analysed. 1µl of whole blood, or dried blood spots soaked in ddH2O, at a dilution of 1 in 2000 were subjected to MALDI-ToF mass spectrometry. MALDI –ToF plates were precoated with 1µl of alpha-cyano-4-hydroxycinnamic acid (CHCA) (Sigma Aldrich) at a concentration of 20mg/ml, dissolved in 1:2 0.1% trifluoroacetic acid and acetonitrile, which was allowed to dry to crystals. 1µl of the diluted patient sample was added and, before completely drying, another 1µl of CHCA was added. After the sample matrix mix had completely dried to crystals, this sample was analysed on the Shimadzu 8020 Clinical Linear MALDI ToF mass spectrometer, which was optimised for analysis at 500 – 700 m/z. MALDI was internally calibrated with singly and doubly charged Bradykinin 1-7 fragment (Sigma Aldrich) protein standard.


The addition of whole blood to ddH2O at a dilution of 1 in 2000 causes red cell lysis, and also dissociation of haemoglobin into free haem and free alpha & beta globins. As these are the most abundant proteins in lysed whole blood by several orders of magnitude, all others molecules are effectively diluted out. Sinapinic acid as the matrix preferentially ionises large proteins and was optimised for globin analysis as previously described (Iles et al. 2013). Lysed whole blood was examined by UV visible spectroscopy. A control sample was also analysed (one male with normal Hb). Three of the patient’s samples at presentation, immediately prior to transfusion, and post transfusion, were analysed. The expected absorption pattern is two major peaks at 550nm (OxyHb) and 670-680nm (DeoxyHb). Comparison of theadmission sample against the control show a large increase in absorption from 600nm to 700nm, with a minor peak at 620nm (arrowed) (Figure 2). Comparison of the admission, start of transfusion and post transfusion samples show a significant decrease in the absorption at 600-700nm post transfusion. However, a peak at 620nm is still visible in the post transfusion sample. This has decreased from 32% of the normal Hb signal at 530nm to 12%, but this is still double that found in the control samples (5-6%). MALDI-ToF mass spectrometric analysis of patient whole blood on the admission sample identified peaks for the expected haem moiety at 618Da, with further peaks at +32Da, +48Da and +64Da, correlating with sulphur, sulphur monoxide and sulphur dioxide bound to haem.


SulfHb was identified in the blood of a patient who presented with persistently low oxygen saturation on pulse oximetry, normal oxygen partial pressure on arterial blood gas sampling, and repeated error messages returned by arterial blood gas machines from two manufacturers for haemoglobin species present. It illustrates the diagnostic difficulties encountered where the rarely encountered SulfHb is present, and the potential for confusion with MetHb (reviewed by Lu et al., 1998). MALDI-ToF MS successfully identified the presence of sulphur, sulphur monoxide and sulphur dioxide bound to haem moieties.

SulfHb is a rare complication of exposure of haem groups to sulphur. It causes irreversible binding of sulphur to the porphyrin ring of the haem moiety, with resultant cyanosis as SulfHb does not bind oxygen. The clinical presentation is similar to that of MetHb, but it does not respond to treatment with methylene blue or ascorbic acid. However, the presence of SulfHb decreases oxygen affinity of unaffected haemoglobin, with left-shift of the oxygen dissociation curve and improved oxygen delivery to tissues; the converse is true of MetHb. SulfHb thus tends to be associated with milder clinical symptoms than MetHb. Where treatment is felt to be required, exchange transfusion is the intervention of choice as it can reduce the proportion of SulfHb; once binding of sulphur to haemoglobin has occurred, it will last for the 120-day lifespan of the erythrocyte.

To the authors’ knowledge, this is the first report using MALDI-ToF MS in the detection of SulfHb. Many previous cases of SulfHb described in the literature are thought to relate to hydrogen sulphide causing SulfHb, often in the context of bacterial overgrowth in the gastrointestinal tract. MALDI-ToF MS enabled the specific identification of sulphur dioxide rather than hydrogen sulphide bound to haem in this patient’s case, establishing the most likely source of SulfHb in this patient as drugs containing sulphur groups; this patient was taking an undeclared Epsom Salts compound, and the problem resolved rapidly when it was ceased. MALDI-ToF MS is a highly specific and rapid methodology for the investigation of dyshaemoglobin molecules; it should be considered alongside gas chromatography for the definitive diagnosis of SulfHb.


Iles RK Iles JK, Abban T, Docherty SM, Naase, M. (2013) METHOD FOR DETECTING ABNORMALITIES IN
HEMOGLOBIN WO Patent App. PCT/GB2015/052,491
Lu HC, Shih RD, Marcus S, Ruck B, Jennis T. Pseudomethemoglobinemia: A Case Report and Review of Sulfhemoglobinemia.
Arch. Pediatr. Adolesc. Med. 1998;152(8):803–805

Original Publication: