Objective There has been some debate about the modes of transmission of genital gonorrhoea in pre-pubertal children. Our objective was to determine, in a consecutive case series from one hospital, what modes of transmission could be determined.
Methods Retrospective review of cases of genital gonorrhoea in pre-pubertal children managed according to an inter-agency assessment protocol from 2002 to 2008. This assessment protocol included extensive screening for potential sources of infection and early forensic interviewing of children old enough to be interviewed.
Results 153 potential contacts of 10 prepubertal children with genital gonorrhoea diagnosed on the basis of positive cultures were investigated. Thirteen persons screened (8.5%) were positive for gonorrhoea, so 9/10 cases had at least one positive potential “contact”. In four children, the mode of transmission was definitely sexual. In four other children, although the source of infection was unclear, there was no evidence of non-sexual transmission. One child probably acquired infection from another child. In one case, the parents argued strongly for non-sexual transmission, but the court regarded this as unproven.
Conclusion Thorough review and contact tracing of pre-pubertal children with genital gonorrhoea found that sexual abuse could be determined as the mode of transmission for at least 40% of children. Although our sample size was limited, we found no case where non-sexual transmission could be determined.
Statistics from Altmetric.com
The incidence of gonorrhoea in New Zealand is increasing, in adults1 and children.2 US guidelines suggest that genital gonorrhoea in a child should be regarded as diagnostic of sexual abuse “if not likely to be perinatally acquired and rare non-sexual transmission is excluded”.3 Earlier British guidelines suggested a “strong probability” of abuse.4 It has been suggested, with regard to some of these guidelines at least, that they “are opinion not evidence-based”,5 lead to denial of the possibility of non-sexual transmission and create a risk of harm and injustice through false accusations of child abuse.5,–,7 This concern appears to be particularly related to one case which presented in Auckland, New Zealand.8 However, a recent evidence-based review concluded that “sexual abuse is the most likely mode of transmission”.9
The purpose of this study was to describe current practice in the management of genital gonorrhoea in children in Auckland and to review cases seen since 2002, with particular reference to the evidence for non-sexual transmission.
What is already known on this topic
▶ Genital gonorrhoea in children is rare and is generally regarded as a sexually transmitted infection.
▶ Historically, fomite transmission was often held responsible for institutional epidemics of paediatric vaginitis attributed to gonorrhoea.
▶ There is very little contemporary evidence for fomite transmission of genital gonorrhoea in children.
What this study adds
▶ We describe a rigorous method for attempting to determine the source and mode of transmission in childhood gonorrhoea.
▶ Using this method, there was no convincing evidence for fomite transmission in a New Zealand cohort.
▶ Even with a rigorous approach to investigation, the source of infection may remain undetermined.
The child protection team of Starship Children's Hospital serves metropolitan Auckland (population 1.4 million)10 and is located with Police and Child Youth and Family (CYF, the statutory child protection agency) in a centre beside the hospital.
We identified all children with genital gonorrhoea seen since 2002,2 reviewed medical records and sought information from CYF, the Police and the Crown Prosecutor.
Management of a prepubertal child with gonorrhoea
Auckland laboratories which identify gonorrhoea in a child (except a neonate) recommend that the doctor consult with our service. We offer to see the child on the day we are consulted. A medical history is taken from the primary caregivers, including details of any suggested non-sexual transmission (such as use of nappies, bed-sharing, bathing and linen) and recent antibiotics. We ask school-aged children about symptoms but do not interview them about how the infection was acquired. A paediatrician with expertise in genital anatomy examines the child.
Confirmation of the infection is sought, by culture of the vagina and (where tolerated), throat and rectum. At the bedside, a direct Gram-stain is prepared, vaginal swabs plated onto New York City media (Fort Richard Laboratories, Auckland, New Zealand) and throat and rectal swabs onto Thayer Martin media (Fort Richard Laboratories). Samples go immediately under chain of evidence to the Regional Sexually Transmitted Infection (STI) laboratory. Neisseria gonorrhoeae is identified from culture by multiple techniques: Gram-stain, oxidase test, RapID NH system (Remel, UK); carbohydrate fermentation (Becton Dickinson, Franklin Lakes, New Jersey, USA) and ProbeTec Strand Displacement Amplification (SDA; Beckton Dickinson). Isolates are stock-cultured at −70°C to allow for later typing.
We look for other STI. If possible, dacron-tipped wire-shafted chlamydia culture swabs are taken from the same sites at the same time and transported in sucrose-phosphate medium (Chlamydia Transwab; Medical Wire and Equipment Company, Corsham, UK). If not processed within 24 h, specimens are stored at −70°C until cultured with standard techniques for chlamydia.11
If tolerated, we culture for Trichomonas vaginalis (wire-shafted swabs, in modified Diamond's TYM media; Fort Richard Laboratories) and other pathogens (Vaginalis Agar media; Microbiology Lab, LabPlus, Auckland, New Zealand). Serology is taken for Syphilis, Hepatitis B and HIV.
The diagnosis is explained to the family, pending confirmation by the second set of samples. Treatment is provided and follow-up arranged.
Our guideline requires immediate notification to CYF and the Police, and a case conference with them within 24 h, where gonorrhoea is explained and information shared. A list is compiled (based on the medical history), of those who have been in contact with the child, which may be modified later if other names are identified. A plan is agreed with a time-line and specific responsibilities, including:
▶ reducing any risk of sexual abuse while an investigation proceeds
▶ minimising emotional harm from action required to ensure the child's safety
▶ appropriate and speedy forensic interview, including good preparation, avoiding any intimidation of the child by others and feedback to caregivers
▶ screening those on the contact list.
Statutory personnel conduct videotaped interviews in developmentally competent children (a forensic interview unit is co-located with our team).
CYF has a legal mandate to take action on behalf of children at risk. The social worker must “form a belief” that a child needs care and protection. New Zealand law emphasises the role of families in securing safety and urges that every effort is made to leave children in the care of their families.
Rationale for screening policy
A contact is defined as someone who was with the child in the 3 weeks before onset of symptoms. Most childhood genital gonorrhoea is symptomatic,14,–,17 but literature defining the incubation period is limited. Standard texts quote adult ranges of 2–7 days,18 but there are reports in children of intervals up to 3 weeks.19 20
Our approach to screening contacts is guided by the following considerations: first, asymptomatic infection can occur (rarely in children),21,–,25 so screening occurs regardless of symptoms. Second, the duration of exposure required is unknown. Household and non-household “associates” may transmit infection.24 Third, associates cannot be excluded by age or sex. Finally, basic principles of confidentiality and consent apply.
Occasionally, the incubation period incorporates a large social event (with opportunity for brief contact with many people otherwise minimally associated) or time at school. In these circumstances, on balance, it seems inappropriate to breach privacy and create anxiety by widespread extra-familial screening, without specific evidence to justify it.
Procedure for screening potential contacts
Contacts are screened by nurse specialists, usually on home visits planned in consultation with the Police. The aim is to screen within 72 h (to minimise any risk of treatment before screening), using urine SDA and specific consent for disclosure to the statutory authorities.
Positive contacts come to clinic for confirmation. Older adolescents or adults are usually seen by a sexual health physician, who arranges treatment and follow-up. Confirmation involves repeat urinary SDA, examination and further samples. In children, the procedure is described above. In adolescents and adults, samples are urethral (culture and SDA), vaginal (SDA), cervical (culture and SDA), throat and rectal (culture and SDA).
Comparison of isolates
Isolates are compared by antimicrobial sensitivity patterns including Minimum Inhibitory Concentration, using full agar plate incorporation techniques, after the system of the Australian Gonococcal Surveillance Program until 2004 and then that of the Clinical and Laboratory Standards Institute (Wayne, Pennsylvania, USA). Molecular typing occurs at the Nosocomial Infections Laboratory of the Institute of Environmental Science and Research (ESR), Wellington, New Zealand, using DNA macro-restriction analysis using pulsed-field gel electrophoresis after digestion with rare-cutting restriction endonucleases (originally XbaI endonuclease, but since 2005 SpeI and NheI). The ESR reports isolates as “indistinguishable”, “closely related” (a 2–3 band difference, equating to the isolates' DNA differing at one site), “possibly related” or “distinct”.26
From 2002 to 2008, 10 cases of genital gonorrhoea in prepubertal children were diagnosed by culture. All were female.
All presented with genital symptoms. Nine were diagnosed unexpectedly on a routine swab for clinical vulvovaginitis taken by a general practitioner. One was the friend of an index case, presenting after her friend's diagnosis. None had behavioural changes or had disclosed sexual contact.
Table 1 lists the medical findings. None had anatomical findings diagnostic of abuse. Case 5 also had chlamydia and case 6 genital warts. Children were followed up in clinic for up to 10 months.
Table 2 shows that 153 people were screened invariably, living at more than one address (range 2–16 houses, median 3). Sixty-four (42%) were children or adolescents. Fifty-six per cent were screened within 72 h and 85% within 7 days. Thirteen persons screened were positive for gonorrhoea (a prevalence of 8.5%), and 9/10 cases had at least one positive potential “contact”.
Table 3 describes statutory processes. In all but case 9, CYF concluded that the child was in need of care and protection, on the basis of probable sexual abuse or (in some cases) other factors such as family violence. These other factors played a key role in the CYF decision to remove two children from their families long term (one for several years, one for 7 months). Two children were temporarily placed outside the family (6 weeks; 2 days). In all these cases, regular access to the mother was provided. Six children remained with family. All competent children were interviewed within a week of our medical examination.
Sexual transmission was established in four cases (1, 4, 5 and 6), on the basis of a confession or disclosure plus matching isolates in perpetrator and victim, and (in two cases) additional STI. One case was not prosecuted because the parents chose not to subject their daughter to the court process.
Sexual transmission was regarded as probable in case 8. Although she made no disclosure, her father had been previously imprisoned for sexual abuse of her older half-sister. He died 2 years before. There had been concern that he may have also sexually abused the brothers, one of whom developed major problems with violence, but neither had made a disclosure. The positive brother's isolate was not fully matched, because he had already been treated elsewhere, but his girlfriend's was identical. Case 8 initiated the sexual play with case 9, raising a possibility of behaviour learned from sexual abuse.
Transmission from cases 8 to 9 by sexual play was regarded as probable. Case 9 described a game where case 8 touched her “fanny” on numerous occasions, sometimes in the bath. The most likely mechanism appears to have been digital transfer of infected secretions to the genitalia.
The mode of transmission was not established in the remaining four cases, in which there were no disclosures or confessions. In case 3, no one screened positive. In cases 2 and 10, a single teenager in each household was positive with an identical isolate, and their contact with the child ceased. No one in these three households described hygiene practices likely to transfer pus onto the genitalia of an 8- or 9-year-old child.
In case 7, the family argued strongly that the infection was acquired from fomites—from bathing, bed-sharing or various other items (wash cloths, towels, clothing or bed clothes). Parents and child had gonorrhoea with identical antimicrobial sensitivities. Isolates were indistinguishable in father and child on molecular typing and very closely related in mother and child (a two-band difference). A two-band difference represents a single genetic event and cannot be taken to suggest that the mother was less likely to be the source of the infection.
This case followed a protracted legal process. The Family Court Judge remarked that:
The alternative possible mechanisms of infection advanced on behalf of the parents require some very specific circumstances. I am not satisfied that the evidence from the parents establishes that those specific circumstances occurred or were likely to have occurred, thereby displacing sexual abuse as the most likely cause of infection. Not only does the parents' evidence not establish on the balance of probabilities the existence of the required circumstances; there are also a number of discrepancies in the parents' evidence, which have caused me to doubt their credibility. There is an element of convenience in the way that the parents' explanations have changed and evolved to meet the scenarios required by their expert's evidence.27
This study describes a comprehensive approach to genital gonorrhoea in children and a rigorous attempt to screen for potential sources of infection. It demonstrates that even with such an approach, there will be children in whom the source of infection cannot be determined. However, in those cases with reliable evidence for a mode of transmission, it has almost always been sexual.
This study provides no support for the view that our approach may lead to “a terrible miscarriage of justice”7 or children being removed from their parents “on the basis of a wrongful assumption of sexual abuse”.6 Only 2 of the 10 children described in this study were removed from their parents' care for a significant length of time, and in those cases, risk factors other than gonorrhoea played a very large part. Only 3 of the 10 cases resulted in prosecution (two supported by a confession from the perpetrator), and all were convicted.
This paper is subject to the limitations of retrospective design, and the small number of cases described. However, all cases were managed relatively uniformly according to a standardised process, and our series included all cases diagnosed in a large centre over 6 years. Because all the laboratories sent positive isolates to the regional laboratory for confirmation, we do not believe any relevant cases were missed.
There are also limitations inherent in our screening method. Despite our best efforts, we did not screen all those identified as contacts, within 72 h. Even had we done so, screening may have missed the source of infection, if the incubation period was incorrect, or relevant persons were omitted. A source may become negative before screening—either spontaneously or after antibiotic treatment. The other possibility, a false negative on urine NAAT, is highly unlikely (in a large study with a gonorrheal prevalence of 3.6%, urine SDA was 99.2% sensitive, with a negative predictive value of 99.9%).28 A negative screen cannot exclude any individual with absolute certainty nor does a positive screen necessarily establish the source of infection. “Screening” is only one aspect of a comprehensive investigation. Nevertheless, immediate screening may be of some help in identifying a safe care giver, reducing the risk of harm caused by unnecessarily separating child and family.
Genital gonorrhoea in a child must be taken seriously. It must be confirmed and treated, the safety and well-being of the child protected and every effort made to determine the source of infection. Agencies must work together closely, if these goals are to be achieved. In our view, it is far better for a child and family to go through a process which is speedy, thorough and professional (even if the result is ambiguous), than to endure the confused and dilatory investigations which may otherwise occur.
Although we found no case to support non-sexual transmission of childhood gonorrhoea, this does not of course exclude the possibility that it may occur. A large multi-centre study would be required to evaluate this, with comprehensive agreed standards of practice in all participating centres.
The authors thank the staff of Te Puaruruhau, Police and CYF for their dedicated work, Mike Brokenshire and Anne McCarthy of LabPlus for the microbiology, and Dr Helen Heffernan, ESR, for the molecular typing and its interpretation.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
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.