Serious skin infections in children: a review of admissions to Gisborne Hospital (2006–2007)

Skin and subcutaneous tissue infections are a heterogeneous group of infections predominantly caused by Staphylococcus aureus (S. aureus) and Streptococcus pyogenes (S. pyogenes).1 They are common childhood complaints in primary care, where they are usually adequately treated (see companion skin infection articles in this issue of the New Zealand Medical Journal). However in an increasing number of cases worldwide, failed or delayed outpatient therapy is leading to more severe disease, requiring costly hospitalisations for often invasive treatment.2-5In New Zealand (NZ), the incidence of serious skin infections in children has almost doubled between 1990 and 2007.5 This increasing disease burden results in important health, social and economic consequences (in 2007 the estimated direct hospitalisation costs alone of these infections was NZ$15 million).5 These infections also contribute to ethnic and deprivation-related health inequalities with evidence of worsening disparities over time.5In NZ, serious skin infection rates are known to be highest in Māori and Pacific Island children, children younger than 5 years old, boys, children living in deprived neighbourhoods and urban areas, and Northern districts of the country.5Risk factors for infection have been reported in a number of international studies as household crowding, close skin to skin contact, undernourishment, low socioeconomic status, poor hygiene, shared bathing, sharing of soap, minor skin trauma, eczema, chickenpox, insect bites, scabies, recent seawater contact, and warm humid climates.6-20 While there are a number of recent national and regional reports,21-25 there are no published studies examining these risk factors in the NZ setting.Gisborne (Tairawhiti) is a region of NZ where skin infections present a major challenge to the health system; the incidence of serious skin infections in children is the highest out of all NZ regions, with evidence of significantly greater ethnic disparities.During the period 1990-2007 the observed incidence of infections in the Tairawhiti region was significantly greater than that expected, even after standardising for the high-risk age, ethnicity and deprivation population composition (see this issue of The New Zealand Medical Journal).This study follows on from that work and aimed to further describe the characteristics of serious skin infections in children of the Tairawhiti region to identify any features that might explain the high burden of disease.Methods A retrospective review was undertaken of clinical notes from all children aged 0-14 years admitted overnight to Gisborne Hospital between 1 January 2006 and 31 December 2007 with a principal or additional diagnosis of serious skin infection. The Tairawhiti region is a relatively isolated area of 45 000 people on the East Coast of NZ's North Island. The region experiences a warm year-round climate and is unique for its large Māori population (47.3% of the total population), youthfulness (26.2% of people are aged less than 15 years old),26 and high level of deprivation (the region has the largest proportion of highly-deprived residents in the country).27 Gisborne Hospital is a 120-bed secondary referral centre which provides inpatient and outpatient health services for the region. Cases of serious skin infections were identified using a defined list of skin infection International Classification of Disease Tenth Revision (ICD-10) codes; this definition was developed in earlier work by the authors and the ICD-10 codes are listed in Appendix 1.28 Day cases, overseas visitors, transfers and readmissions within 30 days with the same diagnosis were excluded. The clinical notes of all selected patients were reviewed by one investigator (CO). Information on patient demographics, prioritised ethnicity, social and environmental characteristics, past medical history, clinical findings, precipitating events, progress and outpatient management of the current infection, investigations, inpatient management and outcome were recorded on a standardised data collection form. Levels of socioeconomic deprivation were assigned based on the patient's home address using the New Zealand Deprivation Index (NZDep); a neighbourhood index based on nine variables extracted from census data where NZDep 1 indicates least deprivation and 10 indicates highest deprivation.29 Information was primarily collected from the records of the current admission, but previous admission notes, general practitioner referral letters and computerised investigation results were reviewed if relevant. Raw data were entered into Microsoft Excel® and analysed in EpiInfo™ (version 3.4.3, Centers for Disease Control and Prevention). Confidence intervals for proportions were calculated using the Wald method. Regional Ethics Committee approval was sought and granted for this study. Results There were 161 children with 163 discrete cases of serious skin infection admitted to Gisborne Hospital between 1 January 2006 and 31 December 2007. These 163 cases accounted for 2.8% of the 5876 serious skin infection paediatric admissions to all NZ hospitals over the study period. Appendix 2 provides a detailed breakdown of patient characteristics and Appendix 3 summarises inpatient investigations, management and outcome of cases. Demographics and environmental characteristics—The mean age of patients was 4.64 years with over half of children in the preschool age group. Males accounted for 54% of cases. Eighty-three percent (n=135) of children were Māori, 14% (n=23) were NZ European/Pakeha, 2% (n=4) Pacific, and the remaining 1% (n=1) other ethnicities. Almost half of cases came from households with residents who smoke, solely outside in 36% of cases and both inside and outside in 13% of cases. The mean number of usual household residents was 5.44 people (range 2-11). Forty-one percent of children measured greater or equal to the 90th weight percentile, with mean weight in the 67th percentile. Past medical history—Fifty-six children (34%) had a recorded history of at least one previous skin infection, with a further 47 (29%) having no documentation of this in their notes. In 20 of the 56 children (12% of total) the previous skin infection was serious, requiring hospitalisation. Ten patients (6%) had a potentially significant pre-existing or concurrent medical condition recorded; these included prematurity (4), impaired glucose tolerance (1), behavioural disorders (1), iron deficiency anaemia (2), Downs' Syndrome (1), and juvenile arthritis requiring systemic immunosuppressants (1). Clinical presentation—The two most common subtypes of infection were cellulitis and subcutaneous abscesses accounting for 38% and 36% of cases respectively. A superficial bacterial infection of a pre-existing skin condition such as eczema, scabies or chickenpox was present in 14% of cases, impetigo in 5%, acute lymphadenitis in 4%, and other specified types of skin infection in the remaining 3% of cases. The head, face and neck and the lower limbs were the most frequently involved sites (32% of cases each), followed by the trunk, groin and buttocks (18%), and upper limbs (11%). Multiple site involvement occurred in 7% of children. Predisposing conditions and pre-hospital management—Just over one-third (37%) of children had a recorded history of trauma to the skin in the 2 weeks prior to the development of the infection. These injuries ranged in type and severity; Table 1 details the individual causes of injury and compares the frequency of these to that documented in previous reports on the Wellington and Auckland regions.22,23 The Tairawhiti region had the highest percentage of cases with a preceding injury identified overall. There was some variation in the distribution of individual causes of injury between the regions; Tairawhiti had the greatest proportion of insect bite/sting related trauma, sports injuries and cuts by a sharp object. Table 1. Identified causes of injury in children with trauma-related serious skin infections in the Tairawhiti, Wellington and Auckland regions Cause Tairawhiti (%) 2006-2007 (n=163) Wellington (%) 1996-200322 (n=1199) Auckland (%) 1994-199823 (n=2055) Insect bite/sting Accidental fall Cut by sharp object Animal related injury Struck by person or object Motor vehicle/cycle or pedestrian accident Sports injury Complication of surgical procedure Vaccination related or iatrogenic Other or unspecified 37.7 9.8 24.6 3.3 4.9 0.0 4.9 1.6 0.0 13.1 20.8 15.1 11.3 8.8 8.2 6.9 3.8 2.5 2.5 20.1 30.0 15.0 22.0 1.0 12.0 6.0 2.0 2.0 2.0 8.0 Total % of cases with preceding injury/trauma identified 37.0 13.3 29.0 Forty-two percent of children had a recorded history of a chronic or sub-acute skin pathology preceding the development of infection. These conditions included eczema (16%), school sores (10%), scabies (6%), varicella (4%) and other conditions (6%). Over three-quarters (77%) of children consulted their general practitioner prior to eventual hospital admission with the median duration of skin infection symptoms prior to this consultation found to be 2 days (mean 2.5 days, range <24 hours to >7 days). Forty percent of children who visited their GP had a course of outpatient antibiotics trialled prior to hospitalisation, the remaining 60% were referred for admission immediately. The median duration of skin infection symptoms prior to hospital admission was 2 days (mean 4.0 days) and ranged from less than 24 hours to longer than a week. Table 2 details the health conditions and management preceding infections in both Māori and non-Māori children. While the number of non-Māori children was too small to enable statistically valid comparisons, the absolute percentages of each variable are not widely divergent. The largest absolute differences are seen in the number of children with a previous serious skin infection; 13% of Māori children and 7% of non-Māori children, and the proportion of cases where antibiotics were started by the GP; 42% of Māori cases and 60% of non-Māori cases. Little absolute ethnic difference is found in the history of a previous skin infection or skin pathology, the frequency of consulting a GP and the duration of symptoms prior to seeking medical attention. Table 2. Predisposing conditions and pre-hospital management of serious skin infections in 0-14-year-old Māori and non-Māori children in the Tairawhiti region, 2006-2007 Variable Total(%) Māori (n=135) Non-Māori (n=28) ƒ % (95% CI) ƒ % (95%CI) Previous skin infection Yes No Not recorded 56 (34) 60 (37) 47 (29) 47 49 39 35(27.3-43.2) 36(28.7-44.7) 29 9 11 8 32(17.8-50.8) 39(23.5-57.6) 29 Previous serious skin infection Yes No Not recorded 20 (12) 139 (85) 4 (3) 18 113 4 13(8.5-20.2) 84(76.5-89.1) 3 2 26 7(0.9-23.7) 93(76.3-99.1) 0 Skin injury/trauma Yes No Not recorded 61 (37) 94 (58) 8 (5) 53 75 7 39(31.4-47.7) 56(47.1-63.7) 5 8 19 1 29(15.1-47.2) 68(49.2-82.2) 4 Skin pathology† Yes No Not recorded 68 (42) 91 (56) 4 (2) 58 73 4 43(34.9-51.4) 54(45.7-62.2) 3 10 18 0 36(20.6-54.3) 64(45.8-79.3) 0 Duration prior to admission <24 hours 1 day 2 days 3 days 4 days 5 days 6 days ≥7 days Not recorded 24 (15) 23 (14) 28 (17) 25 (15) 12 (7) 5 (3) 4 (3) 25 (15) 17 (11) 20 19 26 20 11 5 4 17 13 15(9.7-21.9) 14(9.1-21.0) 19(13.4-26.8) 15(9.7-21.9) 8(4.5-14.1) 4(1.4-8.6) 3(0.9-7.6) 12(7.9-19.3) 10 4 4 2 5 1 0 0 8 4 14(5.1-32.1) 14(5.1-32.1) 7(0.9-23.7) 18(7.4-36.1)) 4(<0.01-19.2) 0 0 29(15.1-47.2) 14(5.1-32.1) Mean/median (days) 3.96/2 3.81/2 4.71/3 Consulted general practitioner Yes No 126 (77) 37 (23) 106 29 79(70.8-84.7) 21(15.4-29.2) 20 8 71(52.8-84.9) 29(15.1-47.2) Duration prior to consulting GP <24 hours 1 day 2 days 3 days 4 days 5 days 6 days ≥7 days Not recorded Not applicable 23 (14) 25 (15) 23 (14) 17 (10) 9 (6) 3 (2) 2 (1) 8 (5) 16 (10) 37 (23) 20 22 19 14 8 2 2 8 11 29 15(9.7-21.9) 16(11.0-23.5) 14(9.1-21.0) 10(6.2-16.8) 6(2.9-11.4) 2(<0.1-5.6) 2(&