Sage Journals: Discover world-class research

Abstract

Yaws is a non-venereal endemic treponemal infection caused by Treponema pallidum sub-species pertenue, a spirochaete bacterium closely related to Treponema pallidum ssp. pallidum, the agent of venereal syphilis. Yaws is a chronic, relapsing disease predominantly affecting children living in certain tropical regions. It spreads by skin-to-skin contact and, like syphilis, occurs in distinct clinical stages. It causes lesions of the skin, mucous membranes and bones which, without treatment, can become chronic and destructive. Treponema pallidum ssp. pertenue, like its sexually-transmitted counterpart, is exquisitely sensitive to penicillin. Infection with yaws or syphilis results in reactive treponemal serology and there is no widely available test to distinguish between these infections. Thus, migration of people from yaws-endemic areas to developed countries may present clinicians with diagnostic dilemmas. We review the epidemiology, clinical presentation and treatment of yaws.

Keywords

Syphilis Yaws non-venereal endemic syphilis neglected tropical diseases

Yaws

Yaws is a non-venereal endemic treponemal infection caused by Treponema pallidum sub-species pertenue,¹ a bacterium closely related to Treponema pallidum ssp. pallidum, the agent of venereal syphilis. Yaws predominantly affects children living in tropical regions of the world. It causes lesions of the skin, mucous membranes and bones which, without treatment, can become chronic and destructive. There is no widely available test to distinguish yaws from syphilis. Thus, migration of people from yaws-endemic areas to developed countries may present clinicians with diagnostic dilemmas. The other endemic treponemal infections are bejel (endemic syphilis) caused by Treponema pallidum ssp. endemicum and pinta caused by Treponema carateum.

Epidemiology

Yaws is currently thought to be endemic in at least 12 countries^2,3 (Table 1). The number of notified yaws cases is almost certainly an underestimate of true disease incidence. Yaws primarily affects children living in poor, densely-populated rural areas. The concentration of yaws in warm, humid climates is thought to be explained by the sensitivity of T p pertenue to relative cool and dryness and may explain why skin lesions are seen more often in the rainy season.⁴

Table 1.

Countries in which yaws is currently endemic.

Country	Local name
Benin	Not Known
Cameroon	Not Known
Central African Republic	Not Known
Democratic Republic of the Congo	Not Known
Congo	Not Known
Côte d’Ivoire*	Goundou
Ghana	Gyator
Togo	Gbodo, Gbodokui,
Indonesia	Frambusia
Timor-Leste	Not Known
Papua New Guinea	Not Known
Solomon Islands	Yaws
Vanuatu	• 50 vatu soa • bigfella soa

*Endemic status unclear.

In the 1950s it was estimated that 50 million people were infected with yaws. The World Health Organization (WHO) tried to eliminate the disease through a mass treatment campaign using benzylpenicillin.^2,5 Consequently, the number of infections worldwide dropped significantly but yaws then fell off the public health agenda. The next 30 years saw a resurgence of cases and the disease is, again, a public health problem in Africa,⁶ South-East Asia, the Pacific^7,8 and South America.⁹ The WHO estimates that 2.5 million individuals may currently be infected.² A failure to identify contacts of infected individuals, inadequate treatment of latent yaws as well as a failure to integrate control efforts into primary health care are thought to have led to the eventual failure of the WHO elimination strategy.⁵

Transmission

Bacteria from infectious lesions enter via a breach in the skin. Lesions of early yaws are most infectious as they carry a higher bacterial load, whilst late yaws lesions are not infectious. It is estimated that infectivity lasts for 12–18 months after primary infection¹ but relapsing disease can extend this period (see ‘latency’ below). It has been postulated that infection might be spread by flies¹⁰ but there is no evidence to support this mode of transmission in humans. Transplacental spread of T p pertenue is said not to occur, but this view is disputed.¹¹

Bacteriology

T p pertenue is a Gram-negative spirochaete which cannot be cultured in vitro.¹ Five strains have been cultured in rabbits and golden hamsters.¹² The organism is closely related to T p pallidum with a genome that differs by approximately 0.2%. These differences are restricted to a small number of genes including tpr and TP0136. The role of these genes is uncertain but they have been implicated in pathogenesis.¹² The phylogenetic relationship of yaws and syphilis remains unclear and there is evidence that recombination between the two organisms can occur.¹³

Clinical presentation

The clinical presentation of yaws bears similarities to that of syphilis (Table 2). Like syphilis, yaws can be staged as early (primary and secondary) and late, or tertiary. Though clinically useful, this classification is artificial and patients may present with a mixture of clinical signs.

Table 2.

Comparison of clinical features and timing of yaws and syphilis.

	Syphilis		Yaws
Primary	Incubation	9–90 days	Incubation	10–90 days
	Morphology	Chancre. Usually solitary, often multiple. Non-tender. Scarring very unusual.	Morphology	Mother yaw. Usually solitary. Non-tender. Scarring usual.
	Site	Ano-genital	Site	Legs, ankles
Secondary	Incubation	Weeks-24 months	Incubation	Weeks-24 months
Secondary	Clinical presentation	Skin rash Lymphadenopathy Mucosal lesions	Clinical presentation	Arthralgia Malaise Skin lesions Polyosteitis of fingers, feet or long bones
Latency	Yes		Yes
Infectious relapses	Commonest within the first two years, rarely thereafter		Up to 5 years, Rarely up to 10 years.
Tertiary	Clinical		Clinical
	Cardiovascular (10%)	Decades	??Cardiovascular	5 + years
	Neurosyphilis (10%)	Weeks (meningitis, cranial neuritis) Decades: tabes, GPI	??Neuroyaws
	Gummata	10–15 years	Gummatous nodules. Scarring, contractures. Gangosa. Tibial bowing. Goundou
Congenital infection	Yes		No evidence

Primary yaws

A papule appears at the inoculation site after about 21 days (range 9–90).^1,10 This ‘Mother Yaw’ may evolve either into an exudative papilloma, 2–5 cm in size or degenerate to form a single, non-tender ulcer (Figures 1 –3) covered by a yellow crust. The legs and ankles are the commonest sites affected, but lesions may occur on the face, buttocks, arms or hands.¹⁴ ‘Split-papules’ may occur at the angle of the mouth.¹ Regional lymphadenopathy is common. In contrast to syphilis, genital lesions are rare. Primary lesions are indolent and take 3–6 months to heal, more often leaving a pigmented scar.¹⁵ As in syphilis,¹⁶ the primary lesion is still present when signs of secondary yaws develop in about 9–15% of patients.¹⁷

Figure 1.

Ulcer of primary yaws. Copyright Michael Marks.

Figure 2.

Ulcer of primary yaws. Copyright Michael Marks.

Figure 3.

Papilloma of primary yaws. Copyright Oriol Mitjà.

Secondary yaws

Haematogenous and lymphatic spread of treponemes produces secondary lesions, most commonly one to two months (but up to 24 months) after the primary lesion. General malaise and lymphadenopathy may occur. The most florid manifestations of secondary yaws occur in skin and bone.¹⁴

Skin

The rash begins as pinhead-size papules, which develop a pustular or crusted appearance and may persist for weeks. If the crust is removed a raspberry-like appearance may be revealed. Sometimes papules enlarge and coalesce into cauliflower-like lesions, most frequently on the face, trunk, genitalia and buttocks. Scaly macules may be seen (Figures 4 and 5). Lesions in warm, moist areas may resemble condylomata lata of syphilis.

Figure 4.

Secondary yaws: multiple small ulcerative lesions. Copyright Michael Marks.

Figure 5.

Secondary yaws: maculo-papular lesions with scaling. Copyright Oriol Mitjà.

The skin lesions of early yaws are often itchy and the Koebner phenomenon has been observed. Mixed papular and macular lesions are often seen in individual patients. Secondary skin lesions may heal even without treatment, with or without scarring.

Squamous macular or plantar yaws can resemble secondary syphilis.¹ Lesions on the soles of the feet may become hyperkeratotic, cracked, discoloured or secondarily infected. This can result in pain and a crab-like gait.¹⁸ Mucous membrane involvement, most commonly nasal, was reported in less than 0.5% of cases in American Samoa.¹⁹

There is some evidence that the manifestations of yaws in the modern era are less florid than previously reported. It has been postulated that use of penicillins to treat other conditions may be responsible for this. The differential diagnosis of yaws lesions is wide and includes syphilis, leishmaniasis, leprosy and Buruli ulcer, as well as non-infectious causes. Discussion with a physician with expertise in tropical medicine is recommended as the differential diagnosis and choice of investigations will vary depending on the patient’s country of origin.

Bones

Secondary yaws typically causes osteoperiostitis of multiple bones. Involvement of long bones may cause nocturnal pain and visible periosteal thickening (Figures 6 and 7). Involvement of the proximal phalanges of the fingers manifests as polydactylitis. This contrasts with late yaws in which mono-dactylitis is typical. One study from Papua New Guinea¹⁴ reported joint pains in 75% of children with secondary yaws.

Figure 6.

Secondary yaws: dactylitis. Copyright Oriol Mitjà.

Figure 7.

Secondary yaws: radiographic evidence of osteoperiostitis. Copyright Oriol Mitjà.

Latent yaws

Individuals with latent yaws have reactive serological tests but no clinical signs. It is not known how many patients are infected without developing clinical disease. Patients with primary and secondary yaws may pass into a period of latency after resolution of clinical signs. As in syphilis, infectious relapses can occur, most commonly up to five years (rarely up to 10 years) after infection.^1,20 Relapsing lesions tend to occur around the axillae, anus and mouth.

Tertiary yaws

Tertiary yaws is thought to occur in about 10% of untreated patients, although its manifestations are rare in the modern era. The skin is most commonly affected. Hyperkeratosis of palms and soles and plaques may occur. Nodules may form near joints and ulcerate, causing tissue necrosis.⁴ ‘Sabre tibia’ results from chronic osteo-periostitis. Gangosa or rhinopharyngitis mutilans denotes mutilating facial ulceration of the palate and nasopharynx secondary to osteitis. Goundou was a rare complication even when yaws was hyperendemic and is characterised by exostoses of the maxillary bones.²¹

Cardiovascular yaws

Although the consensus is that yaws does not cause cardiovascular disease, this view has been challenged. Post-mortem studies have found evidence of aortitis in patients with yaws.²² Histologically these lesions are similar to those found in tertiary syphilis. Despite these studies, definitive evidence of cardiovascular disease in yaws is lacking.

Neurological yaws

The consensus that yaws does not cause neurological disease¹ has also been challenged by studies that found neuro-ophthalmic²³ and CSF abnormalities²⁴ in patients with yaws. As with cardiovascular disease definitive evidence for a causal role of yaws in neurological disease remains absent.

Yaws and pregnancy

While there is no laboratory evidence that T pallidum ssp. pertenue can cause congenital yaws, there are anecdotal reports.¹¹ Most were published when serodiagnosis relied on non-treponemal tests and before treponemal IgM testing of neonates was feasible.¹⁹

Yaws and HIV

There are no published data on the interaction between HIV and yaws. It is possible that patients with latent yaws might develop relapsing disease with increasing immune damage.²⁵ There are also no data on the impact on other STIs, although given the low rates of genital lesions and that the disease predominantly occurs in children it might be anticipated that any effect would be minimal.

Diagnosis

Syphilis or yaws?

Physicians working in endemic areas usually make a presumptive diagnosis of yaws based on clinical and epidemiological features, with or without confirmatory blood tests. However, because syphilis and yaws co-exist in many tropical regions, and serology cannot distinguish between treponemal sub-species, it may be impossible to identify with certainty the causative organism. There are reports of yaws presenting in non-endemic countries.^26,27

Laboratory diagnosis

Dark ground microscopy

Spirochaetes were first observed in yaws ulcers in 1905,²⁸ the year in which T pallidum ssp. pallidum was identified in a lymph gland of a patient with syphilis. T pallidum ssp. pertenue is morphologically identical to T pallidum ssp. pallidum. As T pallidum spp. are only 0.3 µm wide and 6–20 µm in length, dark ground microscopy is required for visualisation. Samples from primary and secondary yaws lesions are obtained as described for syphilis.

Polymerase chain reaction

Polymerase chain reaction (PCR) testing of samples can identify T pallidum but current PCR protocols do not distinguish between sub-species.^14,29 T pallidum ssp. pertenue has been identified to sub-species level using real-time PCR and DNA sequencing in a child from Congo with a pruritic skin eruption,²⁷ but few clinicians have access to such techniques.

Serology

While serological tests are the bedrock of yaws diagnosis they cannot distinguish between sub-species of T pallidum.³⁰

Non-treponemal (cardiolipin) tests

The venereal disease research laboratory (VDRL) and rapid plasma reagin (RPR) tests use an antigen of cardiolipin, lecithin and cholesterol. Patient-derived antibodies produced against lipid in the cell surface of T pallidum react with antigen to cause visible flocculation. The VDRL is read microscopically whereas the RPR can be read with the naked eye. Although non-specific, VDRL/RPR titres best reflect disease activity. Titres fall after treatment and may become zero, especially after treatment of early infection.³¹ RPR titres are generally higher in primary than secondary yaws.¹

Treponemal tests

These include the T pallidum haemagglutination (TPHA) and the T pallidum particle agglutination (TPPA) tests. They are more specific than cardiolipin tests and usually remain positive after treatment.

Point-of-care tests have proved useful in syphilis and results of an initial study in Papua New Guinea suggest they may also be of value in the diagnosis of yaws with good sensitivity and specificity.³² Further studies of these tests in yaws are in progress.

Histology

In early yaws there is marked epidermal hyperplasia and papillomatosis, often with focal spongiosis.³³ Neutrophils accumulate in the epidermis, causing microabscesses. A dense dermal infiltrate of plasma cells is seen.³⁴ In contrast with syphilis, there is little endothelial cell proliferation or vascular obliteration.³⁴ T pallidum can be identified in tissue sections using Warthin-Starry or Levaditi silver stains. While T pallidum ssp. pertenue is found mainly in the epidermis, T pallidum ssp. pallidum is identified more in the dermis.³⁵ Direct and indirect immunofluorescence and immunoperoxidase tests using specific polyclonal antibodies to T pallidum can also be used with histology specimens.³⁶

Radiology

Bone involvement may be revealed by radiographs even when clinical signs are absent (Figure 7).³⁷

Treatment

Benzathine penicillin-G has been the mainstay of treatment for yaws for over 60 years. Lower doses are used compared to syphilis with a recommended dose of 0.6 MU for children (under 10) and 1.2 MU for older children and adults. In a recent single-centre randomised controlled trial, one dose of azithromycin 30 mg/kg was shown to be equivalent to penicillin in patients with primary and secondary yaws, with a cure rate of approximately 95%.³⁸ No other treatment strategies are supported by randomised controlled trials although data from case series suggest oral penicillin can be successful.⁹

Based on these findings, azithromycin is now central to the WHO eradication plan for yaws, which aims to employ community mass treatment in endemic regions. WHO plan to have no further cases of active yaws worldwide by 2017 and to confirm eradication by 2020.^39,40 Despite this optimism there are several barriers to a successful eradication programme including a lack of accurate epidemiological data from many countries where yaws is reported, the absence of dedicated funding for eradication efforts and a concern that resistance to azithromycin, well described in syphilis,⁴¹ will emerge in yaws. Monitoring for this during the eradication programme will be essential. This ambitious plan will require considerable input from NGOs, academic institutions and policy makers.

Response to treatment

Treponemes disappear from lesions within 8–10 hours of treatment with penicillin. Skin lesions begin to heal within 2–4 weeks (Figure 8). In patients with secondary yaws, joint pains may begin to improve within as little as 48 hours.⁴² Bone changes are reversible if treated early enough. Following successful treatment the RPR declines and at 12 months up to 90% of individuals have either a four-fold reduction in RPR or become seronegative.⁴³ Failure of skin lesions to heal or the RPR to drop should be considered treatment failure and an indication for repeat treatment. In endemic settings treatment failure is more common in individuals from higher prevalence communities.⁴⁴ Whether this represents true treatment failure or re-infection is unclear.

Figure 8.

Primary yaws: healed Lesion. Copyright Michael Marks.

The authors of a study in Papua New Guinea reported failure of yaws treatment with penicillin, which they attributed to bacterial resistance, although no laboratory evidence of this was available.⁴⁵

Conclusions

Yaws is still endemic in a number of countries worldwide despite a significant reduction in the number of affected individuals following mass treatment campaigns in the middle of the twentieth century. Clinicians need to be aware of the epidemiology and manifestations of yaws, which should be considered in the differential diagnosis of patients with reactive serology from endemic countries. Older individuals may have acquired yaws in countries that are no longer endemic. Routine testing cannot distinguish between syphilis and yaws. Treatment strategies are similar for the two diseases, although a lower dose of penicillin is used in yaws. Given the limitations in distinguishing the two diagnoses clinicians should consider treating for venereal syphilis in patients with reactive serology without a clear history of yaws. In this context it is important that the clinician carefully explains to the patient and their partner that reactive serology alone is not diagnostic of a sexually-transmitted route of infection.

Development of near-patient and laboratory tests specific for treponemal sub-species is long overdue. We also need to know if yaws can be transmitted from mother to child in utero and whether it can produce neurological and/or cardiovascular complications. Given the prevalence of macrolide and azalide resistance reported in T pallidum ssp. pallidum, it is important that surveillance of treatment efficacy is maintained in planned yaws mass treatment campaigns.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: MM is supported by a Wellcome Trust Clinical Research Fellowship - WT102807. AWS is a Wellcome Trust Intermediate Clinical Fellow (098521) at the London School of Hygiene & Tropical Medicine.

References

Perine PL, Hopkins DR, Niemel PLA, et al. Handbook of endemic treponematoses: yaws, endemic syphilis and pinta, http://apps.who.int/iris/handle/10665/37178?locale=en (1984, accessed 2 May 2013).

WHO | Yaws. WHO, http://www.who.int/mediacentre/factsheets/fs316/en/ (accessed 10 Jun 2013)..

Kazadi

Asiedu

Agana

. Epidemiology of yaws: an update. Clin Epidemiol 2014; 6: 119–128.

Hackett

. Extent and nature of the yaws problem in Africa. Bull World Health Organ 1953; 8: 127–182.

Antal

Causse

. The control of endemic treponematoses. Rev Infect Dis 1985; 7: S220–S226.

Manirakiza

Boas

Beyam

. Clinical outcome of skin yaws lesions after treatment with benzathine benzylpenicillin in a pygmy population in Lobaye, Central African Republic. BMC Res Notes 2011; 4: 543–543.

Dos Santos

Amaral

Harmen

. The prevalence of common skin infections in four districts in Timor-Leste: a cross sectional survey. BMC Infect Dis 2010; 10: 61–61.

Fegan

Glennon

Thami

. Resurgence of yaws in Tanna, Vanuatu: time for a new approach? Trop Doct 2010; 40: 68–69.

Scolnik

Aronson

Lovinsky

. Efficacy of a targeted, oral penicillin-based yaws control program among children living in rural South America. Clin Infect Dis Off Publ Infect Dis Soc Am 2003; 36: 1232–1238.

10.

Castellani

. Experimental investigations on framboesia tropica (yaws). J Hyg (Lond) 1907; 7: 558–569.

11.

Engelhardt

. A study of yaws (does congenital yaws occur?). J Trop Med Hyg 1959; 62: 238–240.

12.

Mikalová

Strouhal

Čejková

. Genome analysis of Treponema pallidum subsp. pallidum and subsp. pertenue strains: most of the genetic differences are localized in six regions. PloS One 2010; 5: e15713–e15713.

13.

Pětrošová

Zobaníková

Čejková

. Whole genome sequence of Treponema pallidum ssp. pallidum, strain Mexico A, suggests recombination between yaws and syphilis strains. PLoS Negl Trop Dis 2012; 6: e1832–e1832.

14.

Mitjà

Hays

Lelngei

. Challenges in recognition and diagnosis of yaws in children in Papua New Guinea. Am J Trop Med Hyg 2011; 85: 113–116.

15.

Sehgal

. Leg ulcers caused by yaws and endemic syphilis. Clin Dermatol 1990; 8: 166–174.

16.

Eccleston

Collins

Higgins

. Primary syphilis. Int J STD AIDS 2008; 19: 145–151.

17.

Powell

. Framboesia: history of its introduction into India; with personal observations of over 200 initial lesions. Proc R Soc Med 1923; 16: 15–42.

18.

Gip

. Yaws revisited. Med J Malaysia 1989; 44: 307–311.

19.

Hunt

Johnson

. Yaws, a study based on over 2,000 cases treated in American Samoa. US Nav Med Bull 1923; 18: 599–607.

20.

Engelkens

Vuzevski

Stolz

. Nonvenereal treponematoses in tropical countries. Clin Dermatol 1999; 17: 143–152. discussion 105–106.

21.

Martinez

Mouney

. Treponemal infections of the head and neck. Otolaryngol Clin North Am 1982; 15: 613–620.

22.

Wilson

Mathis

. Epidemiology and pathology of yaws: a report based on a study of one thousand four hundred and twenty-three consecutive cases in Haiti. J Am Med Assoc 1930; 94: 1289–1292.

23.

Smith

. Neuro-ophthalmological study of late yaws. I. An introduction to yaws. Br J Vener Dis 1971; 47: 223–225.

24.

Hewer

. Some observations on yaws and syphilis in the Southern Sudan. Trans R Soc Trop Med Hyg 1934; 27: 593–608.

25.

Hay

Tam

Kitchen

. Gummatous lesions in men infected with human immunodeficiency virus and syphilis. Genitourin Med 1990; 66: 374–379.

26.

Daly

Morton

. Clinically active yaws in Sheffield. Br J Vener Dis 1963; 39: 98–100.

27.

Pillay

Chen

C-Y

Reynolds

. Laboratory-confirmed case of yaws in a 10-year-old boy from the Republic of the Congo. J Clin Microbiol 2011; 49: 4013–4015.

28.

Castellani

. On the presence of spirochaetes in two cases of ulcerated parangi (Yaws). Br Med J 1905; 2: 1280–1280.

29.

Orle

Gates

Martin

. Simultaneous PCR detection of Haemophilus ducreyi, Treponema pallidum, and herpes simplex virus types 1 and 2 from genital ulcers. J Clin Microbiol 1996; 34: 49–54.

30.

De Caprariis

Della-Latta

. Serologic cross-reactivity of syphilis, yaws, and pinta. Am Fam Physician 2013; 87: 80–80.

31.

Romanowski

Sutherland

Fick

. Serologic response to treatment of infectious syphilis. Ann Intern Med 1991; 114: 1005–1009.

32.

Ayove

Houniei

Wangnapi

. Sensitivity and specificity of a rapid point-of-care test for active yaws: a comparative study. Lancet Glob Health 2014; 2: e415–e421.

33.

Williams

. Pathology of yaws especially the relation of yaws to syphilis. Arch Pathol 1935; 20: 596–630.

34.

Hasselmann

. Comparative studies on the histo-pathology of syphilis, yaws, and pinta. Br J Vener Dis 1957; 33: 5–12.

35.

Hovind-Hougen

Birch-Andersen

Jensen

H-JS

. Ultrastructure of cells of Treponema pertenue obtained from experimentally infected hamsters. Acta Pathol Microbiol Scand [B] 1976; 84: 101–108.

36.

Beckett

Bigbec

. Immunoperoxidase localization of Treponema pallidum: its use in formaldehyde-fixed and paraffin-embedded tissue sections. Arch Pathol Lab Med 1979; 103: 135–138.

37.

Mitjà

Hays

Ipai

. Osteoperiostitis in early yaws: case series and literature review. Clin Infect Dis 2011; 52: 771–774.

38.

Mitjà

Hays

Ipai

. Single-dose azithromycin versus benzathine benzylpenicillin for treatment of yaws in children in Papua New Guinea: an open-label, non-inferiority, randomised trial. Lancet 2012; 379: 342–347.

39.

The World Health Organisation. Eradication of yaws – the Morges Strategy. Wkly Epidemiol Rec 2012; 87: 189–194.

40.

Maurice

. Neglected tropical diseases. Oral antibiotic raises hopes of eradicating yaws. Science 2014; 344: 142–142.

41.

Lukehart

Godornes

Molini

. Macrolide resistance in Treponema pallidum in the United States and Ireland. N Engl J Med 2004; 351: 154–158.

42.

Rein

. Treatment of yaws in the Haitian peasant. J Natl Med Assoc 1949; 41: 60–65.

43.

Soebekti

. Serological study of yaws in Java. Bull World Health Organ 1955; 12: 905–943.

44.

Mitjà

Hays

Ipai

. Outcome predictors in treatment of yaws. Emerg Infect Dis 2011; 17: 1083–1085.

45.

Backhouse

Hudson

Hamilton

. Failure of penicillin treatment of yaws on Karkar Island, Papua New Guinea. Am J Trop Med Hyg 1998; 59: 388–392.