Article Text

Download PDFPDF

Ambulatory intravenous ceftriaxone in paediatric A&E: a useful alternative to hospital admission?
  1. Jennifer K Smith1,
  2. Saji Alexander2,
  3. Ed Abrahamson2
  1. 1Royal Hospital for Sick Children, Edinburgh, UK
  2. 2Chelsea & Westminster Hospital, London, UK
  1. Correspondence to Dr Jennifer K Smith, Paediatric ST5, Royal Hospital for Sick Children, 9 Sciennes Rd, Edinburgh, UK; jennifer.k.smith{at}


Background Treatment of children with intravenous ceftriaxone on an ambulatory basis is described. This allows a child to remain at home, but also be reviewed regularly when attending the Emergency Department for antibiotics.

Methods Indications for, and length of, treatment and laboratory parameters were recorded. Also, a survey of children's parents was undertaken to ascertain opinions regarding ambulatory treatment.

Results 36 patients were treated with ambulatory ceftriaxone over 4 months. Indications included fever without focus, tonsillitis, periorbital cellulitis, urinary tract infection, petechial rash and lymphadenitis. Median duration of treatment was 2.3 days. There was no occult bacteraemia but five positive urine cultures. There was one failure of treatment with subsequent admission for alternative intravenous antibiotics.

Conclusions Parental opinion favours ambulatory treatment, with 94% of parents acknowledging they would choose it again in similar circumstances. Cost analysis favours ambulatory treatment based on predicted costs of a similar length of inpatient stay.

  • Paediatrics
View Full Text

Statistics from


There is no doubt that staying in hospital can be a noisy and disruptive experience. Unwell children particularly benefit from familiar carers in a familiar environment; hospital admission immediately forfeits the latter. If children are ‘unwell’ and need regular clinical monitoring, admission to hospital is the safest and most appropriate option. However, clinicians regularly come across situations where it is believed that intravenous antibiotic treatment is appropriate but the child is not ‘unwell’ enough to require admission. Our unit has guidelines (see box 1) supporting ambulatory antibiotics in an otherwise well but febrile child, who may or may not have a focus. The child is based at home with a cannula in situ and attends the Paediatric Emergency Department daily for clinical review and administration of intravenous antibiotics. Ambulatory intravenous antibiotic use has been minimally reported in UK literature; most publications originate from North America where the healthcare structure and bacteraemic prevalence is quite different to that of the UK. Gauthier et al reported treatment of urinary tract infections (UTIs) with ambulatory gentamicin in 2004.1 Seventy-three per cent of 212 initial diagnoses of UTI in the Emergency Department were suitable for ambulatory treatment. Mean duration of treatment was 1.9 days. Four patients were admitted to hospital with complications. Management of moderate to severe cellulitis with ceftriaxone has been reported by the same centre,2 with 80% of patients being successfully treated as outpatients. Importantly, 95% of families rated the experience as ‘very good’ to ‘excellent’. Earlier work described administering intramuscular ceftriaxone to febrile infants age 1–3 months after a full septic screen (providing blood and cerebrospinal parameters met set criteria.)3 Bacterial infection was subsequently proven in blood, urine or stools of 5.4% of patients, who were treated appropriately.

Box 1

Guidelines for management (during this review, 2008)

Petechial rash

  • Well child with fever >38.5 and petechial rash without symptoms suggestive of viral illness eg, URTI, tonsillitis Do FBC, CRP, blood culture, meningococcal PCR and treat with 80 mg/kg intravenous ceftriaxone until blood culture resulted.

  • This does NOT apply to children who are unwell or exhibit purpura.

Fever without focus

  • As per NICE guideline, acknowledging some children are over 5 years but similar management principles exist. Local use of blind empirical antibiotics: ceftriaxone 80 mg/kg pending cultures in well children after a suitable period of observation and involvement of an SpR of Consultant in decision-making process.

Periorbital cellulitis

  • Minor swelling in well children >5 years can be treated with oral antibiotics (flucloxacillin and cefixime) and a follow-up review in A&E.

  • Children <5 years or more advanced swelling/erythema send FBC, CRP, blood cultures and commence ceftriaxone 80 mg/kg and flucloxacillin 25 mg/kg four times a day. Oral flucloxacillin, ambulatory ceftriaxone and daily A&E review at discretion of senior staff. Continue antibiotics for at least 48 h and until local signs have significantly improved.

Urinary tract infection

  • As per NICE guideline ‘UTI in children’ August 2007.


  • Treat symptomatically with analgesia. Consider admission for fluid management. Antibiotics: 10 days of oral cefalexin if assumed to be bacterial in origin; 7 days amoxicillin if also suspect lower respiratory tract infection.


  • Often requires intravenous treatment; consider ultrasound imaging regarding surgical drainage. Intravenous ceftriaxone 80 mg//kg may be given in well children with regular review.

The concern of occult bacteraemia is justified; studies of febrile children report occult bacteraemia rates of 1.6–5%,4–6 although prevalence has fallen since the introduction of the haemophilus and pneumococcal vaccines. Reduced rates of bacteraemia (0.4%) have been recently quoted; the author warns of the need for vigilance against UTIs, which still affect up to 7.6% of febrile children aged 3–36 months.7

Given the reducing incidence of bacteraemia, one might surmise that it is safer than ever before to be allowing children to be treated on an ambulatory basis. The hospital in which this study is based has guidelines for treatment of most childhood infections encountered in the dedicated Paediatric Emergency Department, including when ambulatory options might be permitted (box 1).


  • To record the population of children presenting to the Paediatric Emergency Department treated with ambulatory intravenous ceftriaxone.

  • To look at indications for use and length of treatment.

  • To report blood and urine culture results, as well as white cell count and C reactive protein results.

  • To perform a cost analysis of inpatient versus outpatient treatment.

  • To carry out a parental questionnaire to determine opinion.


A study period of December to March was chosen because of the higher patient load over winter months. All medical staff were asked to prospectively register the hospital number of any child treated with ambulatory ceftriaxone (16 patients). To ensure that the correct patients had been identified, the hospital patient data manager retrospectively generated a list of all attendances within the study period (9216 attendances). Coded diagnoses of fractures and other injuries were excluded. The notes of remaining patients were manually searched for evidence of ambulatory ceftriaxone prescription; this yielded a further 20 patients.

Data were collected from a combination of written and electronic patient notes and prescription charts. The diagnosis recorded was the initial diagnosis made by the attending clinician.

An eight-point parental questionnaire was created and ethical approval was sought and granted by the local Ethics Committee for use (figure 1).

Figure 1

Parental questionnaire.

Each child had a responsible carer's contact number entered when he/she registered at the A&E. Three telephone numbers were no longer in use or had transferred to new customers when the telephone questionnaire was attempted. One child was related to a clinician working in the department and was excluded. This left 32 contactable carers, all of whom were parents.

The hospital finance department was approached for figures relating to paediatric ward bed costs.


Thirty-six patients received ambulatory ceftriaxone during the study period. Age range was 6 months–15 years, mean 3.8 years. Forty-nine per cent of the study population were under 2 years of age. There were six separate diagnoses: fever without focus (FWF) (33%), petechial rash (28%), UTI (17%), lymphadenitis (5%), periorbital cellulitis (14%) and tonsillitis (3%) (figure 1).

Laboratory parameters yielded a wide range of results: white cell count range 4.7–28.8×109/l (mean 12.3×109/l) and C reactive protein range <5–132 mg/l (mean 41). Blood cultures were performed in 35 out of 36 patients. There were two positive blood cultures, both for coagulase-negative staphylococcus, assumed to be a contaminant. Urine culture was performed on 14 (39%) of the 36 patients. There were five pure growth cultures >105/ml. Two children with a diagnosis of fever of unknown origin had positive urine cultures. It seemed that children with UTIs had higher white cell counts and C reactive proteins than those with other pathologies, but this did not reach statistical significance.

The median number of doses of ceftriaxone was 2.3 (range 1–4). One patient had intramuscular doses, the rest intravenous. There was one failure of treatment in a child with lymphadenitis who was subsequently admitted for intravenous benzylpenicillin and flucloxacillin. All children attended as instructed for follow-up.

In terms of follow-up treatment, 13 (36%) children had intravenous antibiotics only. The rest had varying lengths and types of oral antibiotics depending on the pathology (see table 1).

Table 1

Demographics and results

Given the median number of doses of ceftriaxone, it is possible to very crudely compare costs of treatment. A figure of £502 was quoted for the cost of an overnight paediatric bed in a general ward in this Foundation Trust. The cost, therefore, of 2.3 ‘nights in hospital’ may be extrapolated to over £36 000 for the number of patients described in this study. In reality one cannot simply deduce that the only difference in cost is that of the hospital bed alone, as cost of daily A&E attendances alone for ambulatory patients will not be insignificant. Nevertheless, in the present economy it is surely worth noting.

Eleven of the 32 parents questioned had resided in hospital with a child previously. The majority of parents (72%) felt being at home was better than being admitted, 28% felt it was the same. Nineteen of the 32 parents (60%) admitted they were, at times, worried about their child being at home with an indwelling intravenous cannula. Four children (12.5%) had to have cannulas replaced for subsequent doses of ceftriaxone; another child's cannula dislodged at home. Fifty per cent of parents rated coming up to hospital daily as ‘inconvenient.’ Of these, 81% felt that it was not as inconvenient as residing in hospital with their child. One parent felt it was more inconvenient because of travel time associated with daily visits and waiting times once in the Paediatric Emergency Department. Two were unsure. Travel time from home to hospital ranged from 2 min to over 30 min, mean 19 min. However, most parents acknowledged that time of day and traffic conditions affected these times. Fifty per cent of parents arrived by car, 25% by bus, 15% walked and the remainder journeyed by taxi. When asked if they would choose the ambulatory form of treatment in similar circumstances, 30 out of 32 (94%) said that they would. One parent who would not was that of the child who was admitted to hospital with worsening lymphadenitis. The other felt it to be an ‘overly stressful experience’.

Many parents who were in favour of ambulatory treatment commented independently that they would be willing to have ambulatory antibiotics only if their child was medically well enough to be at home. Other independent comments included ‘children get better more quickly at home’ and ‘families are best in their own environment, especially children’.

Three of the 32 parents questioned commented on lengthy waiting times in the A&E department for subsequent doses of ceftriaxone, but felt this was an expected situation that could not be easily prevented.


This is a small study designed to describe and raise awareness of the use of ambulatory antibiotics in paediatric and emergency practice, and to highlight the positive feedback from parents about this treatment option. It is hoped that paediatricians and emergency clinicians will find the reporting of this treatment method of interest, and even inspiration if it is something they have not considered.

The number of cases would inevitably increase with a longer study interval and perhaps a case of true occult bacteraemia would have resulted.

The peak of ambulatory ceftriaxone treatment was in February. This month was also one of the busiest months of attendances overall. It is noted that the junior doctors change jobs in February and the relative ‘safety net’ of treating with ceftriaxone versus not treating at all may have contributed.

All of the pathologies treated in this study have the potential to be bacterial in nature, with consequences. The ‘fever without focus’ category is traditionally the most challenging, and existing literature delivers no convincing consensus on the value of laboratory or clinical markers in aiding diagnosis of viral versus bacterial pathologies. The recent NICE guideline, ‘Managing feverish children under 5 years’ attempted to address issues for primary and secondary care clinicians. For children over 3 months, a traffic light system based on clinical signs exists. Interestingly, if the child is otherwise well but febrile a urine culture is advised for all, but no routine blood tests or x-rays.8

In our FWF population of 12 children (nine of them <5 years), 100% had blood cultures and only 55% had urine dipstix and culture. Returning to the article from earlier this year, UTIs are far more prevalent than bacteraemia; therefore, more urine samples should be tested and fewer blood samples taken. In the present study population, there were five positive urine cultures with pure growth. This is still by far the most significant bacterial result from the present study and one wonders about the other 45% of the FWF population who did not have urine tested.

The petechial rash group are also of interest. Local guidelines clearly categorise children and advise blood tests and ambulatory ceftriaxone for those with fever but no localising signs on examination, who must also be clinically well. One acknowledges that the latter potentially overtreats those with viral illnesses, but that there will be a small number of children with early bacterial infection who benefit.

In the present study group there was one failure of treatment requiring admission. Lymphadenitis is known for its stubborn qualities, particularly if atypical mycobacteria are implicated. There were no other adverse effects noted within the group, although it cannot be concluded from such small numbers that this is an entirely ‘safe’ practice.

The question of antibiotic resistance is once again raised with this report. A study of bacteraemia in 1999 by Haddon et al recognised that empirical intravenous antibiotic treatment may contribute to such a problem.9 They concluded that although their survey of 534 patients aged 3–36 months revealed occult bacteraemia rates of 3–4%, the risk of antibiotic-related resistance and the lack of reliability of white cell counts and C reactive proteins in deciding to treat made empirical ambulatory antibiotics undesirable. The true bacterial growth in the present study is primarily Escherichia coli. Resistance to ceftriaxone is reported as being 0–2.4% from data looking at community acquired UTIs in children.10 11

The issue of cost is interesting and important. The basic calculations provided earlier are intended to illustrate how much it costs to stay in hospital, not necessarily how much is saved by keeping children at home. A far broader comprehension of cost structures and the complex health economics model is required to truly define any overall cost savings.

The most important message is the favourable parental perspective; the questionnaire generated an encouraging set of data in that it is realised whatever thoughts are about ambulatory treatment it cannot progress if unfavourable to caregivers at home. It is acknowledged that only 34% of parents had stayed in hospital previously so were able to reliably contrast both experiences, but that the overwhelming majority would choose this option again. Working daily in varied clinical settings can lead us to forget stress and fear associated with hospitals. It seems even in a busy capital city with traffic and congestion charges parents prefer to have their children at home and journey daily to hospital for antibiotics, if medically safe.

Ambulatory intravenous antibiotics are not without problems. Fifty per cent of parents still rated the treatment as ‘inconvenient.’ They were concerned if their child had an indwelling cannula in situ at home, mostly because of adhesion difficulties or dislodgement. One child's cannula did dislodge at home; fortunately there was no associated haemorrhage. Some children (12.5%) had to have cannulas replaced, a substantial figure if extrapolated to larger numbers. It is difficult to argue whether the same would have happened in hospital, but not unreasonable to imagine that it may. The busy environment of any A&E department is often not conducive to reliably getting a medical review and dose of ceftriaxone as quickly as parents may expect. Interestingly, in the existing literature most children return to a dedicated ambulatory unit rather than the Emergency Department for subsequent doses. There is increasing preference for health provision that is community-based rather than hospital-based. Many paediatric departments have community nursing staff who attempt to deliver at-home treatments previously restricted to inpatients, examples including oncology and cystic fibrosis patients. Although community nurses can administer antibiotics, there is likely to still be requirement for ongoing medical review of these patients and therefore a completely home-based approach is unattainable.


Ambulatory ceftriaxone is a viable alternative to inpatient stay in the correctly chosen subgroup of patients where the pathology is potentially serious and the treatment appropriate for the suspected pathogen(s). UTIs are much more common than occult bacteraemia and deserve due consideration in the differential diagnosis.

Importantly, ambulatory treatment is acceptable to parents, with a large majority expressing the opinion that they would choose it again in similar circumstances despite acknowledgement of some shortcomings.


View Abstract


  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Charing Cross Research Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

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

  • Primary survey
    Geoff Hughes