Article Text

This article has a correction. Please see:

Download PDFPDF

Transcranial Doppler scanning and the assessment of stroke risk in children with haemoglobin sickle cell disease
  1. Colin R Deane1,
  2. David Goss1,
  3. Sandra O’Driscoll2,
  4. Sarah Mellor2,
  5. Keith R E Pohl3,
  6. Moira C Dick2,
  7. Sue E Height2,
  8. David C Rees2
  1. 1
    Vascular Laboratory, King’s College Hospital, Denmark Hill, London, SE5 9RS, UK
  2. 2
    Department of Paediatric Haematology, King’s College London School of Medicine, King’s College Hospital, Denmark Hill, London, SE5 9RS, UK
  3. 3
    Department of Paediatric Neurology, Evelina Children’s Hospital, Guy’s and St Thomas’ Foundation Trust, Lambeth Palace Road, SE1 7EH, UK
  1. Dr David C Rees, Department of Paediatric Haematology, King’s College Hospital, Denmark Hill, London, SE5 9RS, UK; david.rees{at}


Objective : To assess the role of transcranial Doppler (TCD) scanning in assessing the risk of stroke in children with haemoglobin SC (HbSC) disease. TCD scanning has an established role in primary stroke prevention in sickle cell anaemia but its value in HbSC is unknown.

Design : A retrospective audit of routinely performed TCD scans and routinely collected clinical data.

Setting : A paediatric sickle cell clinic in a teaching hospital in south London, UK.

Patients : 46 children with HbSC disease who have undergone routinely performed TCD scans and steady-state blood tests.

Main outcome measures: The time-averaged mean of the maximum velocity (TAMMV) in the middle cerebral artery circulation correlated with clinical and laboratory data.

Results: The mean TAMMV was 94 cm/s, with a 98th centile of 128 cm/s. This is significantly less than the published ranges for HbSS, with a mean reading of 129 cm/s. One child had a stroke at the age of 5 years, when her TAMMV was measured at 146 cm/s.

Conclusions: Further studies are needed to assess stroke risk in HbSC disease, but we suggest that TCD measurements are potentially useful in this condition, and that readings greater than 128 cm/s are abnormally high and warrant further investigation.

View Full Text

Statistics from


Stroke is one of the major complications of paediatric sickle cell disease (SCD) and affects about 6% of children with the condition by the age of 10 years, with a peak incidence between the ages of two and seven. In SCD, a stroke typically results from progressive narrowing of medium and large blood vessels supplying the brain, particularly the middle cerebral and distal internal carotid arteries, and to a lesser extent the anterior cerebral arteries.1 The exact mechanism of this progressive narrowing is unknown, but it might be related to functional nitric oxide deficiency resulting from haemolysis causing free plasma haemoglobin. Stroke incidence is highest in children with sickle cell anaemia (HbSS), and studies have shown that children at increased risk of stroke can be identified by TCD scanning.2 The STOP study showed that the risk of stroke can be reduced by 90% by regular blood transfusions to keep the amount of haemoglobin S (HbS) in the blood below 30%.3 Nearly all the available studies are based on HbSS, with little information on the risks in sickle cell disease caused by haemoglobin SC disease (HbSC).

HbSC is caused by the co-inheritance of HbS (β6 Glu-Val) and HbC (β6 Glu-Lys). It is the second commonest form of sickle cell disease after sickle cell anaemia and accounts for 25–30% of cases. There are thought to be about 3000 people in the UK with HbS, and in parts of West Africa up to 25% are similarly affected.4 The pathophysiology of HbSC disease differs from that of HbSS in that red cell dehydration induced by HbC is important.5 The clinical features overlap with those of HbSS, but are also distinct. Life-threatening complications are less common, with a median life expectancy of 60 for men and 68 for women in the USA.6 Splenic function is preserved for longer in children with HbSC disease, and the risk of infection is thought to be lower than in HbSS.7 Proliferative retinopathy is about ten times more common in HbSC disease than in HbSS, affecting about 30% of patients.8 Cerebrovascular disease and stroke are also reported to occur in HbSC disease, although at a lower frequency than in HbSS. Studies suggest that the life-time risk of stroke is 2–3%,9 and that the age-adjusted incidence is four times less than in HbSS.1 Most of the studies are centred on HbSS and incidentally include small numbers of HbSC patients.

Although the risk of stroke in people with HbSC disease is lower than with HbSS, studies suggest that there it is still significant and is one of the commoner causes of stroke in childhood. However, there is no specific evidence to guide stroke prevention with HbSC. Some guidelines have been produced in the UK and USA recommending that children with sickle cell disease are screened regularly by TCD scanning for indications of a risk of stroke. The term sickle cell disease implies that all genotypes should be screened, including HbSC, although there is no evidence to support this, and the ranges used for HbSS might be inappropriate. However, other guidelines specifically mention screening patients with HbSS and HbS/β0 thalassaemia, without mentioning HbSC. At King’s College Hospital, London it has been our policy to perform annual transcranial Doppler imaging (TCDI) scans on all children with sickle cell disease for the past 4 years. In this paper, we review our single-centre experience of TCDI scanning and stroke incidence in children with HbSC.


Setting and patients

King’s College Hospital is a teaching hospital in south London, UK. Approximately 400 different children with sickle cell disease are seen each year. Seventy-four children with HbSC disease were identified who had been seen in the paediatric sickle cell clinic in the past two years. In all these patients the diagnosis was confirmed by haemoglobin high-performance liquid chromatography. Routinely collected clinical and laboratory data were analysed. One patient with HbSC disease was known to have had a stroke, and her notes were examined to ascertain various parameters before the stroke. All data were collected as part of routine clinical care and were used anonymously. No extra tests were performed and no extra data were collected from patients. The study was approved by the Clinical Effectiveness Department at King’s College Hospital.

TCD measurements

Intracerebral blood velocity was measured according to the protocol used in the STOP study3 in the vascular laboratory at King’s College Hospital. Colour Duplex scanners (Siemens Sequoia and Aspen ultrasound duplex scanners) were used with a 2 MHz transcranial transducer to record the time-averaged maximum velocity (TAMMV) for the middle cerebral artery, anterior cerebral artery, bifurcation, distal internal carotid artery and posterior cerebral artery on both sides. Previous evaluation of TCD imaging versus TCD scanning had shown that both methods gave the same results when used according to strict, established protocols. TCD measurements were first attempted between the ages of two and four years, depending on the cooperativeness of the child.

Assessments in clinic

In general, children were seen every 6–12 months, unless more frequent visits were clinically necessary. Routine blood tests were performed annually according to clinical protocols and included full blood count, lactate dehydrogenase and assessments of renal and hepatic function. Routine history collection and examination were performed. All measurements were performed when the patient was well and had not received a blood transfusion in the preceding 3 months.


Statistical analysis was performed using SPSS (Statistical Package for the Social Sciences (SPSS) software; SPSS Inc., Chicago, Il, USA).


The combined age of the HbSC children in this study was 600 years, and one patient had a history or signs of a stroke, giving a stroke rate of 0.17/100 patient years.


Seventy-two children (38 female) were studied as summarised in Table 1. Of the patients, 62% had not attended an accident and emergency deparment or been an inpatient in the two years before study. The median number of days in hospital was zero, with a maximum of eight days. The median number of accident and emergency attendances was zero, with a maximum of ten in the preceding two years.

Table 1 Summary of study patients

Transcranial Doppler measurements

Fourty-eight (65%) of the children had had TCD measurements performed according to the STOP protocol.3 Fourteen (19%) were under the age of 3 and had not been scanned. The remainder had not been scanned for a variety of reasons, predominantly patient default. The highest reading from a single intracerebral vessel (TAMMV) was recorded, and the results are summarised in Table 1 and Figure 1. The mean TAMMV was 94 cm/s, with a standard deviation of 15, a 95th centile of 122 and a 98th centile of 128.

Figure 1 Frequency distribution of TCD measurement of maximum TAMMV measurements in 47 children with HbSC disease who had no known neurological complications. The distribution was superimposed on the normal distribution curve.

TAMMV was correlated with other markers of severity in SCD: haemoglobin, neutrophil and platelet count, lactate dehydrogenase, percentage HbF and age. The only significant correlation was a positive relationship between platelet counts and TAMMV (R 0.339, P = 0.020) (Figure 2).

Figure 2 Scatter plot of maximum TAMMV readings against platelet count, with linear regression line (R = 0.339, P = 0.020).

Case history

A five-year-old girl presented herself to an accident and emergency department with three days of left-sided weakness, particularly affecting the leg. She was known to have HbSC disease, which was confirmed subsequently by haemoglobin and DNA analysis. There were brisk reflexes and reduced power on the left side, with an absence of movement in the left leg apart from hip flexion. Brain MRI scanning showed an acute infarct in the territory of the right anterior cerebral circulation, with numerous smaller infarcts in the watershed areas between middle and anterior cerebral arteries bilaterally. Magnetic resonance angiography (MRA) showed occlusion of the distal portion of the right anterior cerebral artery and attenuation of the right middle cerebral artery. At presentation, her haemoglobin was 9.7 g/dl. A thrombophilia screen was subsequently performed and was found to be normal. An emergency exchange transfusion was performed, and good neurological recovery occurred over the next few weeks. She has been maintained for 11 years on regular, simple blood transfusions, to maintain a pretransfusion haemoglobin of 9–10 g/dl with greater than 50% haemoglobin A. She has had no further neurological events. Repeated MR imaging has shown no additional infarcts. TCD scanning was performed on the second day of admission, following the exchange transfusion. The peak velocity was 146 cm/s in the right middle cerebral (MCA) artery, with velocity of 126 in the left MCA. Repeat measurements have remained largely unchanged.


The risk of stroke in children with HbSC disease is 50–100 times greater than that of the general paediatric population, which has a stroke rate of 0.0033 per 100 patient years.1 Although there is good evidence on which to base primary and secondary stroke prevention in HbSS, no such data exist for HbSC. Guidelines on screening for stroke in sickle cell disease either make no mention of HbSC or specifically exclude it.10 Several papers include small numbers of children with HbSC as an adjunct to a study of HbSS. Steen studied 26 patients with HbSC using brain MRI and MRA, and found six with extensive abnormalities on the MRI scan but no stenotic or occluded arteries.11 Grueneich identified seven patients with HbSC disease and found that these children had similar neuropsychological problems to children with HbSS.12 Powars described one patient with HbSC, and 22 with HbSS, with infarctive stroke,13 and Ohene-Frempong found three children with HbSC and stroke in the CSSCD study.1 A study by Verlhac included 12 children with HbSC, 44 with HbSS and two with HbS/β thalassaemia. TCD was used to measure velocity in the MCAs; the data for the HbSC and HbS/β thalassaemia patients were combined, with an average velocity of 77 cm/s in the right MCA, which was not significantly different from the control population.14

Platelet count was significantly correlated with TAMMV in our study, explaining about 11% of the variability. This relationship has not been found in HbSS. There was no correlation between haemoglobin, neutrophil count, lactate dehydrogenase, percentage of haemoglobin F and other parameters thought to be important in HbSS. This raises the possibility that platelets are implicated in the pathogenesis of cerebrovascular disease in HbSC and that antiplatelet agents might be of benefit, although clearly this result and implication needs to be confirmed in a larger study.

Our study suggests that the mean TAMMV in HbSC children without overt stroke is 94 cm/s, with a 98th centile of 128 cm/s. This is significantly lower than measurements in HbSS, in which Adams found a mean TAMMV of 129 cm/s in 167 children with no evidence of cerebrovascular disease.15 Interestingly, the one HbSC patient in our study who had a stroke had a TCD TAMMV of 146 cm/s in her right MCA, significantly more than the 98th centile in our study. The STOP study identified sustained maximum velocities of greater than 170 cm/s as conditionally abnormal, and more than 200 cm/s as abnormal requiring regular blood transfusion.3 These criteria would not seem to be appropriate for HbSC and would have failed to identify the stroke patient in this study as being at increased risk of stroke. Although further studies are clearly required, our data suggest that TAMMV greater than 128 cm/s in HbSC patients could indicate the possibility of significant cerebrovascular disease and trigger the need for further investigations, such as MR imaging and angiography and neuropsychometric testing. There is currently no evidence on which to base a programme of primary stroke prevention in HbSC, and TCD measurement is likely to function as a screen for those requiring further investigation. Although the incidence of stroke in HbSC is less than that in HbSS, the risk is still significant and guidelines need to be developed specifically for this condition.

What is already known on this topic

  • Children with sickle cell anaemia (HbSS) have a 300-fold increased risk of stroke.

  • Increased risk of stroke in HbSS can be detected by TCD screening showing increased intracerebral blood velocities.

What this study adds

  • Intracranial blood velocities are lower in HbSC than HbSS.

  • In HbSC, velocities greater than 128 cm/s are abnormal and should be investigated further.


The authors acknowledge the doctors, nurses and other staff involved in the routine care of the patients.


View Abstract


  • Competing interests: None of the authors have any financial or other conflict of interest that could interfere with the conduct and reporting of this study.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd and Royal College of Paediatrics and Child Health