The classification of vascular anomalies has been hampered historically by confusing and imprecise nomenclature. For example, the terms capillary haemangioma, portwine stain and naevus flammeus have all been used interchangeably to describe what should be more precisely referred to as capillary malformation.

In the “Images in Paediatrics” section of ADC, August 2009 (1), Dr Adhisivam describes a 5-year-old boy with Klippel-Trenaunay syndrome. Although the definition of the syndrome is correctly summarised as a triad including capillary malformation (or port wine stain), hemihypertrophy and venous/lymphatic abnormalities, the patient is incorrectly described with ‘a diffuse capillary haemangioma of the right palm extending proximally’.

Vascular malformations and haemangiomas are separate entities, the terms should not be used interchangeably since the clinical characteristics, prognosis and management for both differs considerably (2). A capillary haemangioma is a vascular tumour which enlarges by rapid cellular proliferation whereas a vascular malformation exhibits a normal rate of endothelial cell turnover.

Common ‘infantile’ haemangiomas are typically absent at birth, usually undergo rapid post-natal growth followed by slow involution. In contrast, capillary malformations (syn. portwine stains) are present at birth, grow proportionately or even progress with time, and usually do not involute. Pulsed dye laser is the treatment of choice for capillary malformations, but is not generally indicated for uncomplicated haemangiomas (3), other than specifically in some cases of persistent ulceration or post-involution residual telangectasia.

References 1. Adhisivam B. Images in Paediatrics: A plethoric palm. Arch Dis Child 2009;94:643 2. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982;69:412-20 3. Batta K, Goodyear HM, Moss C, et al. Randomised controlled study of early pulsed dye laser treatment of uncomplicated childhood haemangiomas: results of a 1-year analysis. Lancet 2002;360:521-7

No competing interests

We were interested to read the study by Edwards et al on the predictive value of their paediatric early warning system (PEWS). They clearly identify the challenges inherent in designing triggers that have both a high sensitivity and specificity. We have an alternate understanding of what PEWS can deliver.

We believe that true value of PEWS is as a situation awareness tool rather than a prediction tool for the deteriorating child. In other words, facilitating knowledge of what is happening rather than what’s going to happen. There is overlap in these functions of course but this subtle difference is essential.

Situation awareness (SA) refers to a person’s perception and understanding of their dynamic environment. This awareness and comprehension is critical in making correct decisions that ultimately lead to correct actions in medical care settings.(1)Put more simply it is “knowing what is going on so you can decide what to do”. Tools that aid situation awareness should ideally pick up on environmental cues and trigger appropriate responses. Deciding on the right cues is important for two reasons. Firstly, as the authors demonstrate, to maximise the sensitivity and specificity of the tool. Secondly the tool should have a mix of cues allowing a 360 degree perspective of the problem. Equally important is deciding on appropriate responses, as this has the ability to somewhat counter balance for lack of specificity. Finally, the effectiveness of any tool depends on its implementation; staff must share the vision of why change is required and they need to participate at every step along the way.

It is important to distinguish what the authors have demonstrated from the actual value of a PEW system or the deployment of Rapid Response Teams (RRTs). The study does not make reference to any interventions that occured when teams became aware of abnormal observations and hence the prevention of adverse outcomes. Measuring the benefit of a change such as PEWS must reflect the effects on safety culture and team work as well as the ability to predict deterioration. Similar problems have been encountered during the evaluation of adult early warning scores. These problems have been used to highlight the different methodologies needed to evaluate such quality improvement interventions, rather than multicentre cluster randomised controled trials, as sugested by the authors.(2)

Situation awareness has been a tested part of team training in other high risk areas (e.g. Aviation and Military). Evaluating which cues will improve sensitivity and specificity to predict the deteriorating children is important but getting too focused on this will stop us implementing beneficial systems with real safety spin offs.(3) If PEWS can help the right people to know what is going on then it can only make our practice safer.

(1) M C Wright, J M Taekman, M R Endsley. Objective measures of situation awareness in a simulated medical environment. Quality and Safety in Health Care 2004;13(Supplement 1) :i65-i71; doi:10.1136/qshc.2004.009951

(2) Berwick D. The Science of Improvement. JAMA 2008;299 (10):1182- 1184.

(3) Tucker, K.M., Brewer, T.L., Baker, R.B. et al. Prospective evaluation of a pediatric inpatient early warning scoring system. Journal for Specialists in Pediatric Nursing, 2009; 14(2), 79-85

Thompson and colleagues have shown that commonly recorded vital signs can be used to identify children with serious infections in the pediatric- assessment-unit and that its sensitivity is comparable to more complicated triage systems (1). However they did not take their study to the next logical step of developing a scoring system for triage, using these vital signs.

We have developed such a scoring system and the validation study done concurrently in the UK and India was presented at the last annual meeting of the British Pediatric Association (2). Called the ‘SICK Score’ (an acronym for Signs of Inflammation in Children that can Kill), it uses the physical variables of the systemic inflammatory response syndrome (SIRS) and its continuum - the multiple organ dysfunction syndrome (MODS). Very much like Thompson and colleagues, the parameters used by us were heart rate, respiratory rate, systolic blood pressure (dichotomized as normotensive and hypotensive), temperature, oxygen saturation, capillary refill time and consciousness on the AVPU scale. The regression coefficients (logs of the odds ratio of death) in the development-study cohort, were used as weights for each parameter. Validation was done against mortality.

The area under the ROC curve for the validation study done in UK and India was 84.1% (95% CI: 77.5 to 90.7%) and this is comparable ROC of 77% in the validation study of PRISM 2 (3). PRISM 2 was used as it is available free in the public domain. In an earlier validation study we tested ‘SICK Score against PRISM score in 125 patients who needed to be admitted to the intensive care unit. The area under the ROC was 0.76 using SICK Score and 0.78 using PRISM score. Hosmer-Lemeshow goodness of fit was excellent. (Hosmer-Lemeshow Chi-square = 2.13 (P = 0.3450)) (4). The software for calculation of SICK Score is available at http://jacob.puliyel.com/sick.php. We agree with Thompson and colleagues that vital signs by themselves can be as useful to assess severity of illness in children seeking medical attention.

Shomi Raman DCH Registrar shomieha@yahoo.co.in

Anjima Basumatari MD Senior Registrar anjimadas@rediffmail.com

Jacob Puliyel MRCP M Phil Consultant puliyel@gmail.com

St Stephens Hospital Delhi 110054

References

1. Thompson MJ, Coad N, Harnden A, Mayon-White R, Perera R, Mant D. How well do vital signs identify children with serious infections in paediatric emergency care? Arch Dis Child published 15 July 2009, 10.1136/adc.2009.159095.

2. Gupta MA, Sahni M, Puliyel JM, Rangasami J, Chakrabarti A, Halstead R, Green DA, Puliyel A, Sreenivas V. International collaboration validating SICK score: a non-invasive severity-of-illness assessment. Arch Dis Child 2008;93 (Suppl 1) A10.

3. Chamberlain JM, Patel KM, Pollack MM. The Pediatric risk of hospital admission score: A second-generation severity-of-illness score for pediatric emergency patients. Pediatrics. 2005;115:388-396.

4. Bhal S, Tygai V, Kumar N, Sreenivas V, Puliyel JM. Signs of inflammation in children that can kill (SICK score): Preliminary prospective validation of a new non-invasive measure of severity-of- illness. J Postgrad Med. 2006;52:102-5.

Dear Editor,

There is no doubt that language is contextual. Disorder as a term, while innocuously descriptive to one, may hold a negative connotation for another, especially those afflicted with whatever said disorder.

Regarding brain sex, I agree with the Consensus Statement on Management of Intersex Disorders (CoSMID) that “Structure of the brain is not currently useful for gender assignment.” Quantification and identification of brain sex haven't been at all forthcoming and fruitful on humans. A prodigious amount of neurons and synaptic connections that have to differentiate and develop would appear to, according to my statistical conjecture, render brain sex and other brain attributes variable (and study incredibly difficult) due to the immense complexity of the process. Gender that, then, combines brain sex and social and environmental factors would also at least, in line with brain sex that influences it, be variable. Sex, if one limits it to the genitalia and perhaps secondary sex characteristics, is dimorphic, as most persons get one of two possibilities when differentiation is complete - that is, if one ignores considerations of size and appearance, which some don't, but I do because functionality, to me, is paramount.

The authors of CoSMID appear to be focused on sex. Brain sex has already been ruled out by their statement on brain structure (see previous quote). Gender, in turn, is too variable to be predictable, and if one reads CoSMID, it is obvious it is not their focus. Most of their focus is at the level of the genitalia. As such, I believe sex was the main consideration for the designation "Disorders of Sexual Development" (DSD).

I take issue with the use of "development." I think it should be "differentiation" - that is, Disorders of Sexual Differentiation - as most DSD problems originate at that level. Development would be from puberty onwards. "Sexual development," in this case, may imply brain development because differentiation and development for the brain can be used interchangeably, but, again, I believe the focus was on sex.

I can only imagine, perhaps not even that, how it must feel to have a condition and it be labeled a "disorder." In a medical setting, however, "disorder" need not be stigmatizing or expedite surgical intervention or hormone replacement. I understand that medical professionals have set the stage with past controversial interventions that will cause them to be viewed with suspicion, but on the other hand we must call a spade a spade. From what is understood about sexual differentiation (and development), DSD cannot be anything other than what the term claims - disorders. Substituting "variation" for "disorder" colloquially is of no consequence, besides the assuagement of the afflicted, which admittedly is important, but scientifically, one would have to wonder about adaptation. If the focus were brain sex or gender, this might be an interesting academic and protracted discussion, but since it is sex (for reasons stated earlier), there are really only one of two possibilities, as most other possibilities tend to be dysfunctional or nonfunctional, leading to infertility. An adaptive benefit, even neutrality, seems doubtful. None of these considerations, however, must necessarily lead to medical intervention or management.

Again, I realize the need to be sensitive to the perceptions of the afflicted ("afflicted" probably will also be seen as a wrong choice of word), but I hope with this to somewhat lessen the supposed stigma "disorder" bears.

Sincerely,

M.A.