A Systematic Review of Risk of HIV Transmission Through Biting or Spitting

Implications for Policy

FV Cresswell; J Ellis; J Hartley; CA Sabin; C Orkin; DR Churchill


HIV Medicine. 2018;19(8):532-540. 

In This Article


Search Results and Study Selection

Our literature search found 1357 citations: 1342 via database searches, and 15 from hand searching of conferences and reference lists. Of these, 615 were duplicates, leaving 742 for title or abstract review. A further 710 were removed because they clearly did not meet the inclusion criteria based on information contained in the title or abstract. The remaining 32 articles underwent full-text review, of which 19 were subsequently removed because they met the exclusion criteria (no primary data, n = 13; exposure of interest not described, n = 1; outcome of interest not described, n = 5), leaving 13 articles in the final data set (Figure 1).

Figure 1.

Flowchart illustrating outcomes of search citations.

Study Characteristics and Quality

Of the 13 studies selected, 11 were case reports and two were case series detailing HIV transmission, or absence of HIV transmission, following a biting episode. There were no reported cases of HIV transmission attributable to spitting. Several of the selected studies were published during the 1980s and 1990s prior to the availability of potent ART.

Of the 13 identified articles that reported alleged HIV transmission related to biting, none related to a bite in the UK and none concerned emergency care workers. The reports included information on a total of 23 people bitten by HIV-positive individuals, of whom nine (39%) seroconverted to HIV positivity following the incident and 14 (61%) did not seroconvert (Table 2). Of these, the alleged transmissions occurred between family members (six of nine), in fights involving infliction of serious wounds (three of nine), or as a result of untrained first-aiders placing fingers in the mouth of someone having a seizure (two of nine).

There was significant heterogeneity in the quality of the reports: a minority had a negative baseline HIV test in the person bitten (two of nine) or phylogenetic analysis of viruses (three of nine). Only four cases in total were classified as having high plausibility or confirmation of HIV infection being attributable to the bite.

Highly Plausible or Confirmed Cases of HIV Transmission Following Bites

Vidmar et al.[21] A first aider was bitten on the hand during a seizure by a man with advanced HIV disease. The biter had confirmed blood in his mouth and was on zidovudine monotherapy, his HIV viral load (VL) was not known and he died 13 days after the incident of primary central nervous system (CNS) lymphoma. The first aider had broken skin at the site of the bite and was HIV-negative on the day of the incident. Despite post-exposure prophylaxis (zidovudine 1200 mg once daily), 33 days later the recipient developed an acute illness and antibody seroconversion was confirmed 54 days after the incident. The recipient had no other risk factors for HIV infection identified.

Centers for Disease Control and Prevention.[22] A person sustained multiple bites from an HIV-positive woman who was reported to have bleeding gums, but who had unknown HIV stage, VL and ART status. It is not reported whether the bites resulted in skin breakage. The recipient was confirmed HIV-negative immediately after the attack and seroconverted 6 weeks later, with RNA sequencing confirming that the perpetrator and recipient shared the same viral strain.

Deshpande et al.[23] A father sustained a bite from his HIV-positive son, causing avulsion of the thumb nail and leaving an exposed bleeding nail bed. The father was not screened for HIV at the time of the bite but presented 4 weeks later with a meningoencephalitis and was found to have acute HIV infection. The son had never received ART and had a VL of 17 163 HIV-1 RNA copies/ml in plasma and 2405 copies/mL in saliva. There were no other risk factors for HIV transmission reported. Sequencing revealed 91% homology between perpetrator and donor HIV RNA.

Andreo et al.[24] A mother was bitten by her son in the context of a seizure. The son was subsequently diagnosed with neurotoxoplasmosis and HIV infection. Blood from a bitten tongue was present in the son's mouth at the time of the incident. The mother's wound was deep and required suturing. She was not screened for HIV at the time of the incident but presented 27 days later with fever and was found to be HIV-positive. DNA sequencing demonstrated that viruses from the mother and son belonged to the same HIV-1 quasi-species.

Medium Plausibility of HIV Transmission Following a Bite

Bartholomew and Jones.[25] A 3-year-old child, born to an HIV-negative mother, was bitten by her father who had dental caries and bleeding gums. He was found to be HIV positive 3 years later (CD4 count 4 cells/μL; HIV VL not measured) and died soon afterwards. The child was therefore tested for HIV and found to be HIV positive. No other risk factors were reported. No phylogenetic analysis was undertaken.

Wahn et al.[26] A child was bitten by his brother who died 6 months after the incident and was diagnosed with toxoplasmosis and HIV infection post-mortem (having received HIV-infected blood during prior cardiac surgery). Family members were screened after his death and the child who had sustained the bite was found to be HIV-positive. The bite allegedly did not result in skin breakage and there was no documentation of blood in the biting child's mouth.

Low Plausibility of HIV Transmission Following a Bite

Khajotia.[27] A man alleged that he contracted HIV infection from kissing during which he sustained a bite on the lip with skin breakage. He reported that the lady who bit his lip was a commercial sexual worker, although she was never confirmed to be HIV positive. He was not screened for HIV at the time of the incident but self-reported multiple negative HIV tests in the subsequent 7 months. He was found to be HIV seropositive while undergoing investigation for gastroenteritis 10 months later. He denied any other risk factor for HIV transmission.

Akani et al.[28] During a fight, a woman was bitten on the lip by her HIV-positive relative. The HIV stage and ART history of the perpetrator were not known, nor was it known whether she had blood in her mouth at the time of the incident. The bite resulted in a deep lip wound requiring suturing. The recipient was not tested for HIV at the time of the bite, but was found to be HIV-positive during antenatal screening 1 year later. The recipient self-reported a negative HIV test prior to the bite, self-reported that her husband was HIV-negative and denied other risk factors for HIV infection, although she had been sexually active and fallen pregnant in the interim.

Anonymous.[29] A woman was bitten by her HIV-positive sister during a fight. The perpetrator was known to be HIV positive and had blood in her mouth at the time of the bite, although her HIV stage, VL and ART status at the time of the incident were not reported. It was not reported whether the bite resulted in breakage of the skin. The recipient was not screened for HIV at the time of the bite, but was found to be HIV seropositive on occupational screening 2 years later. She had a documented negative HIV test 2 years prior to the bite and disclosed three sexual partners in the interim, two of whom were reportedly HIV negative but one of whom was untraceable.