BJD
British Journal of Dermatology
C L I N I C A L A N D LA B O R A T O R Y I N V E S T I G A T I O N S
Cutaneous Mycobacterium chelonae infection in Edinburgh and the Lothians, South-East Scotland, U.K.* V.E. Scott-Lang,1 A. Sergeant,1 C.G. Sinclair,1 I.F. Laurenson,2 A. Biswas,3 M.J. Tidman,1 V.R. Doherty,1 G.M. Kavanagh1 and S.A. Holme1 1 3
Department of Dermatology, and 2Scottish Mycobacteria Reference Laboratory, Royal Infirmary of Edinburgh, Edinburgh EH16 4SA, U.K. Department of Pathology, Western General Hospital, Edinburgh EH4 2XU, U.K.
Summary Correspondence Victoria Scott-Lang. E-mail: [email protected]
Accepted for publication 10 February 2014
Funding sources None.
Conflicts of interest None declared *Plain language summary available online DOI 10.1111/bjd.12901
Background We reviewed all cases of Mycobacterium chelonae infection seen in our department between 1 January 2008 and 31 December 2012. Objectives To review the epidemiology, clinical features and management of cutaneous M. chelonae in South-East Scotland, and to compare prevalence data with the rest of Scotland. Methods The Scottish Mycobacteria Reference Laboratory database was searched for all cases of cutaneous mycobacterial infections. Results One hundred and thirty-four cases of cutaneous mycobacterial infection were recorded. Sixty-three were tuberculous; of the remaining 71, M. chelonae was the most common nontuberculous organism (27 cases). National Health Service (NHS) Lothian Health Board was the area with highest incidence in the Scotland (12 cases). Three main groups of patients in the NHS Lothian Health Board contracted M. chelonae: immunosuppressed patients (n = 6); those who had undergone tattooing (n = 4); and others (n = 2). One case is, we believe, the first report of M. chelonae cutaneous infection associated with topical corticosteroid immunosuppression. The majority of patients were treated with clarithromycin monotherapy. Conclusion The most prevalent nontuberculous cutaneous mycobacterial organism in Scotland is M. chelonae. The prevalence of M. chelonae in Edinburgh and the Lothians compared with the rest of Scotland is disproportionately high, possibly owing to increased local awareness and established facilities for mycobacterial studies. Immunosuppression with prednisolone appears to be a major risk factor. The first outbreak of tattoo-related M. chelonae infection in the U.K. has been reported. Clinicians should be aware of mycobacterial cutaneous infection and ensure that diagnostic skin samples are cultured at the optimal temperatures.
What’s already known about this topic?
• •
Mycobacterium chelonae is a rapidly growing environmental organism found in water and soil. Risk factors for infection include immunosuppression, trauma and invasive medical procedures.
What does this study add?
• • •
© 2014 British Association of Dermatologists
This study demonstrates that M. chelonae is the most common nontuberculous cutaneous mycobacterial infection in Scotland. Tattoo parlours are an important source of infection. Mycobacterial infections should be considered when inflammatory lesions occur in newly tattooed skin and in immunocompromised individuals.
British Journal of Dermatology (2014) 171, pp79–89
79
80 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
Mycobacterium chelonae is a rapidly growing, nontuberculous environmental mycobacterium found in water, soil and sewage. It is not transmissible from human to human: cutaneous infection results from environmental exposure. Cutaneous mycobacterial infections are particularly associated with systemic immunosuppressive drugs, although they may also be seen in immunocompetent individuals, usually following penetrating trauma, including tattooing. Patients may present to a variety of medical specialties, including dermatology, general practice, emergency and general medicine, general surgery and paediatrics. As M. chelonae grows best at lower temperatures than tuberculous mycobacteria, it is important to alert the local laboratory to the possibility of cutaneous mycobacterial infection with the specimen request. A fresh skin biopsy should be sent wrapped in a sterile container and cultured at both 28– 30 °C and 35–37 °C to optimize isolation. Mycobacterium chelonae is classified as a rapidly growing mycobacterium, and visible growth on agar may be seen within 1 day. Molecular techniques can also be employed for diagnosis, but are less sensitive than culture. Treatment may be guided by antibiotic sensitivities, and decisions regarding therapy should be made in conjunction with microbiology colleagues. Over the last 5 years, 12 patients have presented to our department with cutaneous eruptions subsequently confirmed to be M. chelonae. These patients comprised two main groups: those with cutaneous infection secondary to immunosuppression, and recently tattooed individuals (whom we have previously reported).1 To date, there have been no other tattoorelated case reports from the U.K. We sought to identify all Scottish cases of cutaneous mycobacterium infection over the previous 10-year period to put our diagnoses into a national perspective, and to summarize their clinical and management features.
Materials and methods A search was made of the Scottish Mycobacteria Reference Laboratory (SMRL) database. All cases of culture-confirmed cutaneous mycobacterial infection in Scotland (estimated population 5 254 800)2 are recorded in the SMRL database (as
70
nontuberculous mycobacterial infections are not notifiable in Scotland). All cases over a 10-year period, from 1 January 2003 to 31 December 2012, were identified. Case notes were retrieved for patients resident in the area covered by National Health Service (NHS) Lothian Health Board, and data relating to patient demographics, medical history, mode of transmission, histopathological findings, microbiological susceptibilities and treatment were recorded. Skin biopsies were sent for histological examination in all cases. Biopsies were divided in two, with one half being formalin-fixed and 4-lm sections stained with haematoxylin and eosin, periodic acid–Schiff with diastase, Gram, Ziehl–Nielsen and Wade–Fite, and the other half examined for acid-fast bacilli (AFB) and sent for microbiological culture on a solid substrate (Lowenstein Jensen slopes) at 30 °C and 37 °C, and in a liquid substrate (Mycobacterial Growth Indicator Tube; Beckton Dickinson, Franklin Lakes, NJ, U.S.A) at 37 °C. Liquid and solid cultures were considered negative if no growth was detected after 6 or 12 weeks of incubation, respectively. When auramine staining showed AFB, direct molecular examination with in-house realtime polymerase chain reaction was undertaken for Mycobacterium tuberculosis complex and nontuberculous mycobacteria. The latter is often less sensitive than microscopy, and both are less sensitive than culture, which remains the gold standard.
Results There were 134 cases of cutaneous mycobacterial infection in Scotland between 2003 and 2012 (Fig. 1). Of the 71 cases of cutaneous nontuberculous mycobacterial infection, the most common species were M. chelonae (n = 27), Mycobacterium marinum (n = 15), Mycobacterium bovis BCG (n = 9) and Mycobacterium avium (n = 6). The incidence of M. chelonae per 100 000 persons was highest in the NHS Lothian Health Board compared with all other health board areas in Scotland; this appears to be disproportionately high (Table 1). More detailed examination of the NHS Lothian Health Board data (population 848 727)² revealed 12 cases of confirmed M. chelonae infection (Table 2). There was an increase in the incidence in this region over the study period, from the
M. tuberculosis
63
M. chelonae
60
M. marinum
Number of cases
M. bovis BCG
50
M. avium M. fortuitum
40 30
M. malmoense
27
M. gordonae M.intracellulare
20 10 0
15
M. kansassi
9
6
M. scrofulaeceum
3
3
2
2
Organism
British Journal of Dermatology (2014) 171, pp79–89
1
1
1
1
M. palustre M. simiae
Fig 1. Mycobacterial isolates from skin samples, Scotland, January 2003–January 2013. © 2014 British Association of Dermatologists
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 81 Table 1 Incidence of Mycobacterium chelonae in Scottish National Health Service (NHS) Health Boards per 100 000 persons from 1 January 2003 to 31 December 2012
NHS Scotland Health Board
Total population within Health Board area
Percentage of Scottish population as a whole
Number of cases of M. chelonae (percentage of all Scottish cases)
Incidence per 100 000 persons
Greater Glasgow and Clyde Lothian Lanarkshire Grampian Tayside Fife Ayrshire and Arran Highland Forth Valley Dumfries and Galloway Borders Western Isles Shetland Orkney
1 210 848 563 555 405 367 366 311 295 148 113 26 22 20
23 16 107 105 77 698 698 59 56 28 25 049 042 038
6 12 0 2 0 3 0 2 0 0 2 0 0 0
033 141 0 036 0 054 0 032 0 0 088 0 0 0
254 727 185 280 721 292 890 960 541 060 150 080 500 160
earliest single confirmed case in 2008 to a maximum of four cases in both 2011 and 2012 (Fig. 2). There were eight women and four men, aged 11–97 years. The infected patients fell into three main groups: the immunocompromised (n = 6), those with tattoo-related infections (n = 4) and others (n = 2). The six systemically immunosuppressed patients (five women and one man) presented in a 4-year period, all with cutaneous M. chelonae infections affecting a limb. All patients were on systemic steroid therapy: four were on systemic steroid monotherapy, and two were on systemic corticosteroids along with other immunosuppressive agents, including rituximab, cyclophosphamide and alemtuzumab. The majority of the immunosuppressed patients were women aged 51–86 years. Following a diagnosis of M. chelonae, one patient was weaned from prednisolone and one patient had leflunomide discontinued: immunosuppressive therapy was unable to be withdrawn in the remaining patients. Two of the immunocompromised patients died: a 51-year-old man with chronic lymphocytic leukaemia, and a 78-year-old woman who had an intracerebral haemorrhage while on antibiotic therapy. None of the patients had evidence of systemic mycobacterial infection. To date, four patients have been confirmed with culturepositive M. chelonae within tattoo sites: none of the tattoorelated patients were immunocompromised. Three separate clusters of tattooed patients have now presented in Edinburgh and the Lothians over the last 2 years, all with erythema and pustulation developing within freshly-applied grey areas within their tattoos. In addition, there have been five clinically and pathologically suspicious tattoo-related cases in which the culture for atypical mycobacterium was negative. The first cluster (one confirmed and three suspected cases) was published in 2012.1 All the individuals had attended the same tattoo parlour in Edinburgh. Subsequent to this, three further patients were seen with similar eruptions within the grey areas © 2014 British Association of Dermatologists
(22) (44) (0) (8) (0) (10) (0) (8) (0) (0) (8) (0) (0) (0)
of tattoos, and had M. chelonae confirmed on skin culture. Two of these three patients attended the same tattoo parlour in West Lothian, and the third patient had been tattooed by a friend. Two further patients, who had attended different tattoo parlours, have also been seen: in these cases there was a high clinical suspicion of M. chelonae, although culture was negative. In all cases, clinicians highlighted to both pathologists and microbiologists the possible diagnosis of atypical mycobacterial infection. All patients were negative for hepatitis B and hepatitis C viruses, and HIV. Following initial identification of the tattoo-related clusters, all patients with tattoo reactions who had been seen and photographed in the previous year were re-reviewed in the dermatology department, leading to identification of one further case of M. chelonae, which is included in this report. All patients, including those with unconfirmed microbiology responded to antimicrobials for mycobacterial infection, helping to substantiate the clinical diagnosis. Plastic surgery colleagues, who are also referred patients with tattoo-related problems, were contacted and asked to consider the diagnosis in any new patients referred with a reaction to a tattoo. The two other cases included one elderly woman with bullous pemphigoid, in whom M. chelonae infection developed on her lower legs at the site of application of clobetasol propionate. To our knowledge, this is the first reported case of M. chelonae developing in the context of topical corticosteroid application. The second patient was an 11-year-old boy who developed M. chelonae on his cheek due to trauma, having fallen in the bathroom at home, striking his face against a tap. In the 12 patients with a diagnosis of cutaneous M. chelonae, the time from biopsy to culture result ranged from 1 to 43 days (mean 12 days). The most common site of infection was the upper arm (54%), followed by the leg (31%). Eleven of the 12 cases were diagnosed by a dermatologist. Clinically, the affected patients typically presented with papules and pustules, indurated erythema, erosions and ulcers (Fig. 3). In the British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology (2014) 171, pp79–89
54 (F)
2
86 (F)
51 (M)
1
3
Age (sex)
Case
Wegener granulomatosis
CLL
PMHx
Yes
Yes
Yes
Immunosuppressed
Prednisolone
Rituximab, cyclophosphamide, prednisolone
Alemtuzumab, dexamethasone
Medication
Leg. Admitted by medics after a fall at
Left lower leg. Noted to have abscess-like lesions on left upper arm and ulcer on left leg in January 2009. Subsequently developed three nodules on the left shin. Pus was aspirated and grew M. chelonae
Right arm. Pustules and erythema originally treated as shingles. Biopsied 5 months later by dermatology department
Distribution and presentation
Intense mixed inflammatory infiltrate within subcutaneous tissue composed of neutrophils and histiocytes with abscess and granuloma formation and necrosis. Acid-fast bacilli seen
Tissue culture positive. Clarithromycin
Scattered dermal chronic
Rifampicin, ethambutol and clarithromycin for 18 months
Dermal collection of mixed inflammatory cells, including prominent macrophages. Microorganisms seen
Tissue culture positive. Clarithromycin
Triple therapy with ethambutol, rifampicin and clarithromycin started when M. chelonae cultured. Ethambutol withdrawn after 10 months. Rifampicin and clarithromycin stopped after 2 years. Restarted clarithromycin and rifampicin 6 months later owing to concerns about relapse following collapse of nasal bridge secondary to Wegener granulomatosis. Plan is to continue lifelong clarithromycin and rifampicin Initially treated with combination of
Treatment
Histology
Culture and sensitivities
Table 2 Summary of 12 cases of Mycobacterium chelonae seen in Edinburgh and the Lothians between 1 January 2008 and 31 December 2012
Completed 6-month course of
Completed 15 months of continuous triple therapy. Skin described as ‘good’ at the time of stopping antibiotics. Patient died 3 months later from concurrent medical issues related to CLL Patient continues on lifelong clarithromycin and rifampicin, and has never developed any further cutaneous lesions
Response to treatment and follow-up
82 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
© 2014 British Association of Dermatologists
© 2014 British Association of Dermatologists
86 (F)
5
78 (F)
85 (F)
4
6
Age (sex)
Case
Brain tumour
Nodular vasculitis
Giant cell arteritis
Polymyalgia Rheumatic arthritis
PMHx
Table 2 (continued)
Yes
Yes
Yes
Immunosuppressed
Dexamethasone
Prednisolone
Prednisolone, leflunomide
Medication
Arm. Patient had been on dexamethasone preoperatively to shrink a brain tumour. Developed erythema, pustules and ulcers on arm near site of repeated attempts at venepuncture
Right arm. Patient on long-term prednisolone for over 20 years. Presented with pustules and nodules on the right forearm
Right arm and hand. Referred to dermatology department with papulopustular eruption on right forearm and hand
home. Referred to dermatology with pustules, nodules and superficial ulcers on left leg
Distribution and presentation
Focal areas of ulceration with dense nodular inflammatory cell infiltrate involving the dermis and subcutaneous fat. There was an admixture of neutrophil-rich suppurative inflammation around ‘pseudocystic’ spaces Tissue culture positive. Tobramycin, amikacin, clarithromycin and linezolid
Tissue culture positive. Clarithromyci and tobramycin
Localized inflammatory infiltrate, vaguely granulomatous in areas with necrosis. Numerous elongated microorganisms identified, particularly on Wade –Fite stain Multiple areas of suppurative granulomatous inflammation Abundant bacillary organisms seen on Wade–Fite and Ziehl –Neelsen staining
inflammatory cell infiltrate with occasional small aggregates of neutrophils. No organisms seen
Histology
Tissue culture positive. Clarithromycin and tobramycin
Tissue culture positive. Clarithromycin
Culture and sensitivities
Took 6-month course of clarithromycin, which resulted in minimal induration and little residual erythema. Repeat biopsy taken off antibiotics was negative on culture Commenced on IV amikacin initially and then tobramycin added in. Patient developed an INR > 10 and subsequent intracerebral haemorrhage. Patient was not fit for neurosurgical intervention, and became increasingly unresponsive and died
Clarithromycin for 6 months. Prednisolone weaned
Amikacin then tobramycin
clarithromycin and rifampicin. Repeat skin biopsy performed when off antibiotics was negative for M. chelonae on culture Completed 3-month course of clarithromycin, which resulted in complete clearance of lesions
Response to treatment and follow-up
clarithromycin, rifampicin and tobramycin. Tobramycin was withdrawn owing to impaired renal function Clarithromycin for 3 months. Leflunomide stopped
Treatment
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 83
British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology (2014) 171, pp79–89
12 (F)
97 (F)
7
8
Age (sex)
Case
Nil
Bullous pemphigoid
PMHx
Table 2 (continued)
No
Not systemically. On superpotent topical steroid
Immunosuppressed
Nil
Clobetasol propionate to legs
Medication
Cheek. Patient fell in bathroom and struck his cheek against the tap. Sustained a laceration that resulted in a nonhealing abscess
Bilateral legs. Developed papules and pustules on lower legs after application of clobetasol propionate for biopsy-confirmed bullous pemphigoid
Distribution and presentation
Fluid from right cheek laceration positive on culture. Clarithromycin and tobramycin
Tissue culture positive. Amikacin, ciprofloxacin, clarithromycin, linezolid and tobramycin
Culture and sensitivities surrounded by poorly formed aggregates of histiocytes. Numerous AFB seen on Wade–Fite staining. PASd and Gram stains were negative Prominent mixed inflammatory focus within the dermis, including foci of abscess formation around a distorted and cystically dilated hair follicle. Infiltrate loosely granulomatous in places, with a few multinucleate giant cells. Numerous mycobacterial organisms identified on the Wade–Fite and Ziehl–Neelson stains Excision of abscess cavity: dermis and subcutaneous tissue showed evidence of extensive, rather nodular, fibrosis. Associated with patchy inflammation and, in places, inflamed granulation tissue. No evidence
Histology
Rifampicin, clarithromycin and ciprofloxacin
Initially treated with topical bactroban and then clarithromycin added
Treatment
Patient completed 6 months of tripletherapy antibiotics. Antibiotics stopped as it was felt that there was no residual infection. Subsequently underwent revision of significant scar,
Skin improved with topical antibiotics and a 2-month course of clarithromycin. No recurrence
Response to treatment and follow-up
84 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
© 2014 British Association of Dermatologists
Age (sex)
33 (M)
30 (M)
42 (M)
Case
9
10
© 2014 British Association of Dermatologists
11
Nil
Nil
Nil
PMHx
Table 2 (continued)
No
No
No
Immunosuppressed
Nil
Nil
Nil
Medication
Arm, tattoo site
Arm, tattoo site
Arm, tattoo site
Distribution and presentation
Tissue culture positive. Amikacin, ciprofloxacin, clarithromycin and tobramycin
Tissue culture positive. Amikacin, ciprofloxacin, clarithromycin, cotrimoxazole, linezolid and tobramycin
Tissue culture positive. Clarithromycin and tobramycin
Culture and sensitivities
Clarithromycin for 4 months
Clarithromycin for 3 months, then ciprofloxacin for 2 weeks, but patient felt unwell. Then, a combination of moxifloxacin and clarithromycin for 3 months Florid mixed inflammation within the superficial dermis, focally granulomatous in nature. No organisms seen
Clarithromycin for 6 months
Treatment
Granulomatous inflammation with a neutrophilic infiltrate within the dermis. No organisms seen
of granulomatous inflammation identified. No organisms seen Granulomatous inflammation with a neutrophilic infiltrate within the dermis. No organisms seen
Histology
Marked improvement following 3 months of clarithromycin. A further 3-month course was recommended, but the patient failed to attend for follow-up appointment Patient prescribed clarithromycin by GP over a 4-month period and reported improvement on this to his GP. He failed to return to dermatology and has not been contactable by telephone Patient took clarithromycin for 3 months, but continued to have areas of induration and papules within the grey area of the tattoo. Was switched to ciprofloxacin, but it was stopped after 2 weeks owing to side-effects. Then had a 3-month course of moxifloxacin and clarithromyin.
with a good cosmetic outcome
Response to treatment and follow-up
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 85
British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology (2014) 171, pp79–89 Immunosuppressed
No
PMHx
Nil
Age (sex)
28 (F)
Nil
Medication
Histology
Foci of suppurating granulomatous inflammation in superficial dermis. Tattoo pigment was identified. No mycobacterial organisms seen
Culture and sensitivities
Tissue culture positive. Tobramycin, amikacin, ciprofloxacin, linezolid, moxifloxacin, clarithromycin
Distribution and presentation
Arm, tattoo site. Developed papular erythematous eruption within grey-shaded areas of new tattoo
3-month course of clarithromycin
Treatment
Repeat biopsy was negative on culture Patient prescribed 3month course of clarithromycin and then failed to attend the dermatology review appointment. Patient contacted by telephone and stated that the tattoo had improved following 2 months of antibiotics
Response to treatment and follow-up
Number of cases
PMHx, patient medical history; M, male; F, female; PASd, periodic acid–Schiff with diastase; CLL, chronic lymphocytic leukaemia; IV, intravenous; INR, international normalized ratio; GP, general practitioner.
12
Case
Table 2 (continued)
86 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
4 Number of cases in NHS Lothian 2003 –2012
3
2
1
0
Year
Fig 2. Incidence of cases of cutaneous Mycobacterium chelonae infection in the NHS Lothian Health Board area between 1 January 2003 and 31 December 2012.
(a)
(b)
Fig 3. (a, b) Patients with Mycobacterium chelonae infection typically presented with papules and pustules, indurated erythema, erosions and ulcers.
tattooed patients, all inflammatory eruptions were confined to areas of grey shading (Fig. 4). Histologically, a suppurative granulomatous dermatitis/panniculitis was the predominant
© 2014 British Association of Dermatologists
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 87
(a)
(b)
Fig 4. (a, b) All inflammatory eruptions in tattooed patients were confined to areas of grey shading.
pattern of inflammation, occurring in seven of the 12 patients. A purely tuberculoid granulomatous pattern was found in three patients. The remaining two patients showed an abscesslike acute inflammatory picture with granulation tissue-like small vessel proliferation. AFB were detected in four patients (all immunosuppressed), all of whom presented with suppurative granulomatous inflammation with pseudocyst-like empty spaces. None of the tattoo-related cases showed any demonstrable microorganisms on special histochemical stains.
Discussion Mycobacterium chelonae is a ubiquitous environmental organism, and our experience suggests that cases of cutaneous infection appear to be increasing. Previously identified sources of M. chelonae include tap water, water tanks and interventional medical equipment, such as catheters and bronchoscopes. In recent years, infection has also been reported as a complication of cosmetic procedures, including injection of botulinum toxin,3 contaminated foot spas4 and liposuction.5 Outbreaks have also been seen following mesotherapy6 and laser eye surgery.7 The largely unregulated nature of the U.K.’s cosmetic industry and the increasing demand for invasive procedures by the general public may be contributing to the apparent increase in incidence. It is well recognized that immunosuppressed individuals are at particular risk of infections, including cutaneous mycobacterial infections. In our series of 12 patients, all of the six immunosuppressed individuals were on oral steroids, and three were on additional systemic immunosuppression, including biological therapy, cyclophosphamide and leflunomide. None of the immunosuppressed individuals were known to have visceral involvement, but there were two deaths from concur© 2014 British Association of Dermatologists
rent medical problems. We have also reported what we believe to be the first case of M. chelonae at the site of clobetasol propionate steroid application. Of additional interest is our documentation of four confirmed cases of M. chelonae secondary to tattooing, and an additional five further suspected cases in which the clinical features, epidemiological factors and histology were highly suggestive of infection, but skin culture was negative, possibly owing, in part, to concurrent antibiotic treatment. To date, these clusters of patients from Edinburgh and the Lothians are the only cases to be reported in the U.K., but other case series from France and the U.S.A. have been published.8–11 All patients in our series with tattoo-related infection had undergone tattooing with grey ink, and all had very similar eruptions comprising red papules, pustules and induration confined to areas of grey shading. The Civic Government (Scotland) Act 1982 (Licensing of Skin Piercing and Tattooing) Order of 2006 legislates the legal requirements for tattoo parlours in Scotland.12 The Act requires tattoo artists to use only sterile, single-use disposable needles for skin piercing and tattooing, and to use only single pigment or ink prepackaged in single-use vials. However, there is no requirement for tattoo artists to use sterile water for rinsing needles, or diluting down black ink to grey, and it is possible that tattoo artists and ink manufacturers may use tap water for this process. In the first reported cluster of patients in Edinburgh, M. chelonae was cultured from an opened bottle of ink and from rinsings of the ink bottle nozzle, but not from unopened bottles, which was suggestive of environmental contamination. The tattoo parlour implicated in the second outbreak in West Lothian admitted to using nonsterile tap water to dilute black ink down to a grey wash, and also used tap water in a rinse pot used for cleaning needles. Authors of other published cases of tattoo-related M. chelonae infection have also postulated that the source of the infection may be the use of nonsterile water to dilute black ink down to grey, either at a manufacturing level or within tattoo parlours. Consequent to the discovery of tattoo-related M. chelonae clusters in south-east Scotland, a communication was issued to all general practitioners (GPs), consultant dermatologists, microbiologists, plastic surgeons and pharmacists working for NHS Lothian Health Board to highlight the problem and to encourage the referral of possible cases. Public and environmental health departments have been in close contact with the relevant tattoo parlours locally. The outbreaks noted in south-east Scotland and in other countries suggest the necessity of establishing national and international guidelines on sterile techniques in tattoo parlours, including the avoidance of nonsterile water, and tightening of the regulations governing the manufacture and supply of tattoo inks. The cases of cutaneous M. chelonae that we report are similar clinically to other reported cases of cutaneous, nontuberculous mycobacterial infections, including those related to M. marinum, Mycobacterium abscessus and Mycobacterium fortuitum. A case series of 51 patients with cutaneous, nontuberculous mycobacterial infection reported that the most commonly observed lesions British Journal of Dermatology (2014) 171, pp79–89
88 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
were papules and pustules; other patients developed abscesses, inflammatory plaques, nodules and suppurative ulceration.13 All of these clinical features were seen in our patients with M. chelonae infection. These clinical findings, particularly in the context of an immunosuppressed individual, should alert the clinician to the possibility of mycobacterial infection. It is noted that no M. abscessus was isolated from skin specimens across Scotland; the reason for this is unclear. Mycobacterium abscessus is said to be a common cause of skin mycobacterial infection;14 however, reports of this infection from environmental sources are scant, and may represent identification methods that discriminate poorly between M. chelonae and M. abscessus. Alternatively, there may be variable geographical distribution and host interactions accounting for the lack of cutaneous M. abscessus infection in Scotland. In contrast to the apparently increasing incidence of nontuberculous mycobacterial cutaneous infections, cutaneous tuberculosis (TB) remains relatively rare within Europe, although the prevalence differs around the world.15 It is postulated that the incidence within Europe will rise with increased immigration from outside the continent. Cutaneous TB appears to differ clinically from nontuberculous mycobacterial infections, and can take a number of recognized patterns, including lupus vulgaris and scrofuloderma (typically the most commonly reported forms in adults), TB verrucosa cutis, miliary TB and tuberculid. The histological spectrum of nontuberculous mycobacterial infection is rather broad and can be similar to M. tuberculosis infection.16 Cutaneous M. chelonae infections (like the other rapid-growing mycobacterial species, e.g. M. fortuitum) tend to have a distinctive appearance characterized by suppurative granulomatous inflammation surrounding pseudocyst-like structures often containing abundant AFB.17 We have demonstrated a striking difference in the incidence of M. chelonae throughout Scotland. It is possible that there is a higher prevalence of M. chelonae in soil and water in south-east Scotland, but it is, perhaps, more likely that, compared with the rest of the country, there is an increased local clinical and laboratory awareness of the possibility of atypical mycobacteria, resulting in appropriate samples being rapidly submitted and cultured optimally at correct temperatures, with an increased diagnostic pick-up. Most Scottish laboratories culture mycobacteria in liquid or solid media only, and do not culture at both 30 °C and 37 °C, relying on the referral of specimens to the SMRL for specialist culture. The tattoo outbreaks are likely not unique to south-east Scotland, and the lack of confirmed cases in other areas is probably multifactorial, such as patients not seeking medical attention for inflamed tattoos, fewer GP referrals of patients with tattoo reactions, empirical antibiotic treatment clouding the clinical picture, misdiagnosis as allergic contact dermatitis, sarcoid or other granulomatous reaction, or a low index of suspicion for mycobacterial infections in the diagnostic differential, resulting in inadequate or inappropriate investigations. Given the risk factors for developing M. chelonae, including immunosuppression, cosmetic procedures and tattooing, it is possible the incidence will further British Journal of Dermatology (2014) 171, pp79–89
increase in the U.K. Therefore, familiarity with the clinical features, diagnosis and management strategies of cutaneous mycobacterial infection is essential for both dermatologists and nondermatologists to allow prompt diagnosis and initiation of therapy. Management may be guided by antibiotic sensitivities and in consultation with microbiology colleagues. Patients are likely to require a prolonged course of antibiotic therapy, usually for a minimum of 3 months. Clarithromycin monotherapy may often suffice, but some patients may require combinations of antibiotics to ensure resolution. A reduction of immunosuppression may need to be considered in some patients. In particular, the diagnosis should considered in patients with eruptions who are taking long-term systemic glucocorticoids, as our series appears to suggest that they are at high risk of infection.
Acknowledgments We thank colleagues from the Scottish Mycobacteria Reference Laboratory (SMRL) for their assistance. Identification and susceptibility testing is the remit of the SMRL, which is funded through Health Protection Scotland, by National Services Scotland.
References 1 Sergeant A, Conaglen P, Laurenson IF et al. Mycobacterium chelonae infection: a complication of tattooing. Clin Exp Dermatol 2013; 38:140–2. 2 National Records of Scotland (NRS). Census results. Available at: http://www.scotlandscensus.gov.uk/en/censusresults/ (last accessed 24 March 2014). 3 Latorre-Gonzalez G, Garcia-Garcia M, Martinez-Colubi M et al. [Disseminated cutaneous infection by Mycobacterium chelonae after botulinic toxin in an immunosuppressed patient]. Med Clin (Barc) 2005; 125:439 (in Spanish). 4 Sniezek P, Graham B, Busch H, Lederman E. Rapidly growing mycobacterial infections after pedicures. Arch Dermatol 2003; 139:629–34. 5 Meyers H, Brown-Elliott BA, Moore D et al. An outbreak of Mycobacterium chelonae infection following liposuction. Clin Infect Dis 2002; 34:1500–7. 6 Munayco C, Grijalva C, Culqui D et al. Outbreak of persistent cutaneous abscesses due to Mycobacterium chelonae after mesotherapy sessions, Lima, Peru. Rev Saude Pu blica 2008; 42:146–9. 7 Freitas D, Alvarenga L, Sampaoio J et al. An outbreak of Mycobacterium chelonae infection after LASIK. Ophthalmology 2003; 110:276–85. 8 Kluger N, Muller C, Gral N. Atypical mycobacteria infection following tattooing: a review of an outbreak in 8 patients in a French tattoo parlour. Arch Dermatol 2008; 144:941–2. 9 Goldman J, Caron F, de Quatrebarbes J et al. Infections from tattooing. Outbreak of Mycobacterium chelonae in France. BMJ 2010; 341: c5483. 10 Drage L, Ecker P, Orenstein R et al. An outbreak of Mycobacterium chelonae infections in tattoos. J Am Acad Dermatol 2010; 62:501–6. 11 Kennedy B, Bedard B, Younge M et al. Outbreak of Mycobacterium chelonae infection associated with tattoo ink. N Engl J Med 2012; 367:1020–4. 12 legislation.gov.uk. The Civic Government (Scotland) Act 1982 (Licensing of Skin Piercing and Tattooing) Order 2006. Available © 2014 British Association of Dermatologists
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 89 at: http://www.legislation.gov.uk/ssi/2006/43/contents/made (last accessed 24 March 2014). 13 Bartralot R, Garcıa-Patos V, Sitjas D et al. Clinical patterns of cutaneous nontuberculous mycobacterial infections. Br J Dermatol 2005; 152:727–34. 14 Griffith D, Aksamit T, Brown-Elliott B et al. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007; 175:367–416.
© 2014 British Association of Dermatologists
15 Marcoval J, Alcaide F. Evolution of cutaneous tuberculosis over the past 30 years in a tertiary referral hospital on the European Mediterranean coast. Clin Exp Dermatol 2012; 38:131–6. 16 Santa CD, Strayer DS. The histologic spectrum of the cutaneous mycobacteriosis. Hum Pathol 1982; 13:485–95. 17 Kriegsman M, Scott-Lang VE, Holme SA, Biswas A. Suppurative and granulomatous dermatitis with pseudocysts: a useful tissue reaction pattern. Diagn Histopathol (Oxf) 2012; 18:185–8.
British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology
C L I N I C A L A N D LA B O R A T O R Y I N V E S T I G A T I O N S
Cutaneous Mycobacterium chelonae infection in Edinburgh and the Lothians, South-East Scotland, U.K.* V.E. Scott-Lang,1 A. Sergeant,1 C.G. Sinclair,1 I.F. Laurenson,2 A. Biswas,3 M.J. Tidman,1 V.R. Doherty,1 G.M. Kavanagh1 and S.A. Holme1 1 3
Department of Dermatology, and 2Scottish Mycobacteria Reference Laboratory, Royal Infirmary of Edinburgh, Edinburgh EH16 4SA, U.K. Department of Pathology, Western General Hospital, Edinburgh EH4 2XU, U.K.
Summary Correspondence Victoria Scott-Lang. E-mail: [email protected]
Accepted for publication 10 February 2014
Funding sources None.
Conflicts of interest None declared *Plain language summary available online DOI 10.1111/bjd.12901
Background We reviewed all cases of Mycobacterium chelonae infection seen in our department between 1 January 2008 and 31 December 2012. Objectives To review the epidemiology, clinical features and management of cutaneous M. chelonae in South-East Scotland, and to compare prevalence data with the rest of Scotland. Methods The Scottish Mycobacteria Reference Laboratory database was searched for all cases of cutaneous mycobacterial infections. Results One hundred and thirty-four cases of cutaneous mycobacterial infection were recorded. Sixty-three were tuberculous; of the remaining 71, M. chelonae was the most common nontuberculous organism (27 cases). National Health Service (NHS) Lothian Health Board was the area with highest incidence in the Scotland (12 cases). Three main groups of patients in the NHS Lothian Health Board contracted M. chelonae: immunosuppressed patients (n = 6); those who had undergone tattooing (n = 4); and others (n = 2). One case is, we believe, the first report of M. chelonae cutaneous infection associated with topical corticosteroid immunosuppression. The majority of patients were treated with clarithromycin monotherapy. Conclusion The most prevalent nontuberculous cutaneous mycobacterial organism in Scotland is M. chelonae. The prevalence of M. chelonae in Edinburgh and the Lothians compared with the rest of Scotland is disproportionately high, possibly owing to increased local awareness and established facilities for mycobacterial studies. Immunosuppression with prednisolone appears to be a major risk factor. The first outbreak of tattoo-related M. chelonae infection in the U.K. has been reported. Clinicians should be aware of mycobacterial cutaneous infection and ensure that diagnostic skin samples are cultured at the optimal temperatures.
What’s already known about this topic?
• •
Mycobacterium chelonae is a rapidly growing environmental organism found in water and soil. Risk factors for infection include immunosuppression, trauma and invasive medical procedures.
What does this study add?
• • •
© 2014 British Association of Dermatologists
This study demonstrates that M. chelonae is the most common nontuberculous cutaneous mycobacterial infection in Scotland. Tattoo parlours are an important source of infection. Mycobacterial infections should be considered when inflammatory lesions occur in newly tattooed skin and in immunocompromised individuals.
British Journal of Dermatology (2014) 171, pp79–89
79
80 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
Mycobacterium chelonae is a rapidly growing, nontuberculous environmental mycobacterium found in water, soil and sewage. It is not transmissible from human to human: cutaneous infection results from environmental exposure. Cutaneous mycobacterial infections are particularly associated with systemic immunosuppressive drugs, although they may also be seen in immunocompetent individuals, usually following penetrating trauma, including tattooing. Patients may present to a variety of medical specialties, including dermatology, general practice, emergency and general medicine, general surgery and paediatrics. As M. chelonae grows best at lower temperatures than tuberculous mycobacteria, it is important to alert the local laboratory to the possibility of cutaneous mycobacterial infection with the specimen request. A fresh skin biopsy should be sent wrapped in a sterile container and cultured at both 28– 30 °C and 35–37 °C to optimize isolation. Mycobacterium chelonae is classified as a rapidly growing mycobacterium, and visible growth on agar may be seen within 1 day. Molecular techniques can also be employed for diagnosis, but are less sensitive than culture. Treatment may be guided by antibiotic sensitivities, and decisions regarding therapy should be made in conjunction with microbiology colleagues. Over the last 5 years, 12 patients have presented to our department with cutaneous eruptions subsequently confirmed to be M. chelonae. These patients comprised two main groups: those with cutaneous infection secondary to immunosuppression, and recently tattooed individuals (whom we have previously reported).1 To date, there have been no other tattoorelated case reports from the U.K. We sought to identify all Scottish cases of cutaneous mycobacterium infection over the previous 10-year period to put our diagnoses into a national perspective, and to summarize their clinical and management features.
Materials and methods A search was made of the Scottish Mycobacteria Reference Laboratory (SMRL) database. All cases of culture-confirmed cutaneous mycobacterial infection in Scotland (estimated population 5 254 800)2 are recorded in the SMRL database (as
70
nontuberculous mycobacterial infections are not notifiable in Scotland). All cases over a 10-year period, from 1 January 2003 to 31 December 2012, were identified. Case notes were retrieved for patients resident in the area covered by National Health Service (NHS) Lothian Health Board, and data relating to patient demographics, medical history, mode of transmission, histopathological findings, microbiological susceptibilities and treatment were recorded. Skin biopsies were sent for histological examination in all cases. Biopsies were divided in two, with one half being formalin-fixed and 4-lm sections stained with haematoxylin and eosin, periodic acid–Schiff with diastase, Gram, Ziehl–Nielsen and Wade–Fite, and the other half examined for acid-fast bacilli (AFB) and sent for microbiological culture on a solid substrate (Lowenstein Jensen slopes) at 30 °C and 37 °C, and in a liquid substrate (Mycobacterial Growth Indicator Tube; Beckton Dickinson, Franklin Lakes, NJ, U.S.A) at 37 °C. Liquid and solid cultures were considered negative if no growth was detected after 6 or 12 weeks of incubation, respectively. When auramine staining showed AFB, direct molecular examination with in-house realtime polymerase chain reaction was undertaken for Mycobacterium tuberculosis complex and nontuberculous mycobacteria. The latter is often less sensitive than microscopy, and both are less sensitive than culture, which remains the gold standard.
Results There were 134 cases of cutaneous mycobacterial infection in Scotland between 2003 and 2012 (Fig. 1). Of the 71 cases of cutaneous nontuberculous mycobacterial infection, the most common species were M. chelonae (n = 27), Mycobacterium marinum (n = 15), Mycobacterium bovis BCG (n = 9) and Mycobacterium avium (n = 6). The incidence of M. chelonae per 100 000 persons was highest in the NHS Lothian Health Board compared with all other health board areas in Scotland; this appears to be disproportionately high (Table 1). More detailed examination of the NHS Lothian Health Board data (population 848 727)² revealed 12 cases of confirmed M. chelonae infection (Table 2). There was an increase in the incidence in this region over the study period, from the
M. tuberculosis
63
M. chelonae
60
M. marinum
Number of cases
M. bovis BCG
50
M. avium M. fortuitum
40 30
M. malmoense
27
M. gordonae M.intracellulare
20 10 0
15
M. kansassi
9
6
M. scrofulaeceum
3
3
2
2
Organism
British Journal of Dermatology (2014) 171, pp79–89
1
1
1
1
M. palustre M. simiae
Fig 1. Mycobacterial isolates from skin samples, Scotland, January 2003–January 2013. © 2014 British Association of Dermatologists
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 81 Table 1 Incidence of Mycobacterium chelonae in Scottish National Health Service (NHS) Health Boards per 100 000 persons from 1 January 2003 to 31 December 2012
NHS Scotland Health Board
Total population within Health Board area
Percentage of Scottish population as a whole
Number of cases of M. chelonae (percentage of all Scottish cases)
Incidence per 100 000 persons
Greater Glasgow and Clyde Lothian Lanarkshire Grampian Tayside Fife Ayrshire and Arran Highland Forth Valley Dumfries and Galloway Borders Western Isles Shetland Orkney
1 210 848 563 555 405 367 366 311 295 148 113 26 22 20
23 16 107 105 77 698 698 59 56 28 25 049 042 038
6 12 0 2 0 3 0 2 0 0 2 0 0 0
033 141 0 036 0 054 0 032 0 0 088 0 0 0
254 727 185 280 721 292 890 960 541 060 150 080 500 160
earliest single confirmed case in 2008 to a maximum of four cases in both 2011 and 2012 (Fig. 2). There were eight women and four men, aged 11–97 years. The infected patients fell into three main groups: the immunocompromised (n = 6), those with tattoo-related infections (n = 4) and others (n = 2). The six systemically immunosuppressed patients (five women and one man) presented in a 4-year period, all with cutaneous M. chelonae infections affecting a limb. All patients were on systemic steroid therapy: four were on systemic steroid monotherapy, and two were on systemic corticosteroids along with other immunosuppressive agents, including rituximab, cyclophosphamide and alemtuzumab. The majority of the immunosuppressed patients were women aged 51–86 years. Following a diagnosis of M. chelonae, one patient was weaned from prednisolone and one patient had leflunomide discontinued: immunosuppressive therapy was unable to be withdrawn in the remaining patients. Two of the immunocompromised patients died: a 51-year-old man with chronic lymphocytic leukaemia, and a 78-year-old woman who had an intracerebral haemorrhage while on antibiotic therapy. None of the patients had evidence of systemic mycobacterial infection. To date, four patients have been confirmed with culturepositive M. chelonae within tattoo sites: none of the tattoorelated patients were immunocompromised. Three separate clusters of tattooed patients have now presented in Edinburgh and the Lothians over the last 2 years, all with erythema and pustulation developing within freshly-applied grey areas within their tattoos. In addition, there have been five clinically and pathologically suspicious tattoo-related cases in which the culture for atypical mycobacterium was negative. The first cluster (one confirmed and three suspected cases) was published in 2012.1 All the individuals had attended the same tattoo parlour in Edinburgh. Subsequent to this, three further patients were seen with similar eruptions within the grey areas © 2014 British Association of Dermatologists
(22) (44) (0) (8) (0) (10) (0) (8) (0) (0) (8) (0) (0) (0)
of tattoos, and had M. chelonae confirmed on skin culture. Two of these three patients attended the same tattoo parlour in West Lothian, and the third patient had been tattooed by a friend. Two further patients, who had attended different tattoo parlours, have also been seen: in these cases there was a high clinical suspicion of M. chelonae, although culture was negative. In all cases, clinicians highlighted to both pathologists and microbiologists the possible diagnosis of atypical mycobacterial infection. All patients were negative for hepatitis B and hepatitis C viruses, and HIV. Following initial identification of the tattoo-related clusters, all patients with tattoo reactions who had been seen and photographed in the previous year were re-reviewed in the dermatology department, leading to identification of one further case of M. chelonae, which is included in this report. All patients, including those with unconfirmed microbiology responded to antimicrobials for mycobacterial infection, helping to substantiate the clinical diagnosis. Plastic surgery colleagues, who are also referred patients with tattoo-related problems, were contacted and asked to consider the diagnosis in any new patients referred with a reaction to a tattoo. The two other cases included one elderly woman with bullous pemphigoid, in whom M. chelonae infection developed on her lower legs at the site of application of clobetasol propionate. To our knowledge, this is the first reported case of M. chelonae developing in the context of topical corticosteroid application. The second patient was an 11-year-old boy who developed M. chelonae on his cheek due to trauma, having fallen in the bathroom at home, striking his face against a tap. In the 12 patients with a diagnosis of cutaneous M. chelonae, the time from biopsy to culture result ranged from 1 to 43 days (mean 12 days). The most common site of infection was the upper arm (54%), followed by the leg (31%). Eleven of the 12 cases were diagnosed by a dermatologist. Clinically, the affected patients typically presented with papules and pustules, indurated erythema, erosions and ulcers (Fig. 3). In the British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology (2014) 171, pp79–89
54 (F)
2
86 (F)
51 (M)
1
3
Age (sex)
Case
Wegener granulomatosis
CLL
PMHx
Yes
Yes
Yes
Immunosuppressed
Prednisolone
Rituximab, cyclophosphamide, prednisolone
Alemtuzumab, dexamethasone
Medication
Leg. Admitted by medics after a fall at
Left lower leg. Noted to have abscess-like lesions on left upper arm and ulcer on left leg in January 2009. Subsequently developed three nodules on the left shin. Pus was aspirated and grew M. chelonae
Right arm. Pustules and erythema originally treated as shingles. Biopsied 5 months later by dermatology department
Distribution and presentation
Intense mixed inflammatory infiltrate within subcutaneous tissue composed of neutrophils and histiocytes with abscess and granuloma formation and necrosis. Acid-fast bacilli seen
Tissue culture positive. Clarithromycin
Scattered dermal chronic
Rifampicin, ethambutol and clarithromycin for 18 months
Dermal collection of mixed inflammatory cells, including prominent macrophages. Microorganisms seen
Tissue culture positive. Clarithromycin
Triple therapy with ethambutol, rifampicin and clarithromycin started when M. chelonae cultured. Ethambutol withdrawn after 10 months. Rifampicin and clarithromycin stopped after 2 years. Restarted clarithromycin and rifampicin 6 months later owing to concerns about relapse following collapse of nasal bridge secondary to Wegener granulomatosis. Plan is to continue lifelong clarithromycin and rifampicin Initially treated with combination of
Treatment
Histology
Culture and sensitivities
Table 2 Summary of 12 cases of Mycobacterium chelonae seen in Edinburgh and the Lothians between 1 January 2008 and 31 December 2012
Completed 6-month course of
Completed 15 months of continuous triple therapy. Skin described as ‘good’ at the time of stopping antibiotics. Patient died 3 months later from concurrent medical issues related to CLL Patient continues on lifelong clarithromycin and rifampicin, and has never developed any further cutaneous lesions
Response to treatment and follow-up
82 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
© 2014 British Association of Dermatologists
© 2014 British Association of Dermatologists
86 (F)
5
78 (F)
85 (F)
4
6
Age (sex)
Case
Brain tumour
Nodular vasculitis
Giant cell arteritis
Polymyalgia Rheumatic arthritis
PMHx
Table 2 (continued)
Yes
Yes
Yes
Immunosuppressed
Dexamethasone
Prednisolone
Prednisolone, leflunomide
Medication
Arm. Patient had been on dexamethasone preoperatively to shrink a brain tumour. Developed erythema, pustules and ulcers on arm near site of repeated attempts at venepuncture
Right arm. Patient on long-term prednisolone for over 20 years. Presented with pustules and nodules on the right forearm
Right arm and hand. Referred to dermatology department with papulopustular eruption on right forearm and hand
home. Referred to dermatology with pustules, nodules and superficial ulcers on left leg
Distribution and presentation
Focal areas of ulceration with dense nodular inflammatory cell infiltrate involving the dermis and subcutaneous fat. There was an admixture of neutrophil-rich suppurative inflammation around ‘pseudocystic’ spaces Tissue culture positive. Tobramycin, amikacin, clarithromycin and linezolid
Tissue culture positive. Clarithromyci and tobramycin
Localized inflammatory infiltrate, vaguely granulomatous in areas with necrosis. Numerous elongated microorganisms identified, particularly on Wade –Fite stain Multiple areas of suppurative granulomatous inflammation Abundant bacillary organisms seen on Wade–Fite and Ziehl –Neelsen staining
inflammatory cell infiltrate with occasional small aggregates of neutrophils. No organisms seen
Histology
Tissue culture positive. Clarithromycin and tobramycin
Tissue culture positive. Clarithromycin
Culture and sensitivities
Took 6-month course of clarithromycin, which resulted in minimal induration and little residual erythema. Repeat biopsy taken off antibiotics was negative on culture Commenced on IV amikacin initially and then tobramycin added in. Patient developed an INR > 10 and subsequent intracerebral haemorrhage. Patient was not fit for neurosurgical intervention, and became increasingly unresponsive and died
Clarithromycin for 6 months. Prednisolone weaned
Amikacin then tobramycin
clarithromycin and rifampicin. Repeat skin biopsy performed when off antibiotics was negative for M. chelonae on culture Completed 3-month course of clarithromycin, which resulted in complete clearance of lesions
Response to treatment and follow-up
clarithromycin, rifampicin and tobramycin. Tobramycin was withdrawn owing to impaired renal function Clarithromycin for 3 months. Leflunomide stopped
Treatment
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 83
British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology (2014) 171, pp79–89
12 (F)
97 (F)
7
8
Age (sex)
Case
Nil
Bullous pemphigoid
PMHx
Table 2 (continued)
No
Not systemically. On superpotent topical steroid
Immunosuppressed
Nil
Clobetasol propionate to legs
Medication
Cheek. Patient fell in bathroom and struck his cheek against the tap. Sustained a laceration that resulted in a nonhealing abscess
Bilateral legs. Developed papules and pustules on lower legs after application of clobetasol propionate for biopsy-confirmed bullous pemphigoid
Distribution and presentation
Fluid from right cheek laceration positive on culture. Clarithromycin and tobramycin
Tissue culture positive. Amikacin, ciprofloxacin, clarithromycin, linezolid and tobramycin
Culture and sensitivities surrounded by poorly formed aggregates of histiocytes. Numerous AFB seen on Wade–Fite staining. PASd and Gram stains were negative Prominent mixed inflammatory focus within the dermis, including foci of abscess formation around a distorted and cystically dilated hair follicle. Infiltrate loosely granulomatous in places, with a few multinucleate giant cells. Numerous mycobacterial organisms identified on the Wade–Fite and Ziehl–Neelson stains Excision of abscess cavity: dermis and subcutaneous tissue showed evidence of extensive, rather nodular, fibrosis. Associated with patchy inflammation and, in places, inflamed granulation tissue. No evidence
Histology
Rifampicin, clarithromycin and ciprofloxacin
Initially treated with topical bactroban and then clarithromycin added
Treatment
Patient completed 6 months of tripletherapy antibiotics. Antibiotics stopped as it was felt that there was no residual infection. Subsequently underwent revision of significant scar,
Skin improved with topical antibiotics and a 2-month course of clarithromycin. No recurrence
Response to treatment and follow-up
84 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
© 2014 British Association of Dermatologists
Age (sex)
33 (M)
30 (M)
42 (M)
Case
9
10
© 2014 British Association of Dermatologists
11
Nil
Nil
Nil
PMHx
Table 2 (continued)
No
No
No
Immunosuppressed
Nil
Nil
Nil
Medication
Arm, tattoo site
Arm, tattoo site
Arm, tattoo site
Distribution and presentation
Tissue culture positive. Amikacin, ciprofloxacin, clarithromycin and tobramycin
Tissue culture positive. Amikacin, ciprofloxacin, clarithromycin, cotrimoxazole, linezolid and tobramycin
Tissue culture positive. Clarithromycin and tobramycin
Culture and sensitivities
Clarithromycin for 4 months
Clarithromycin for 3 months, then ciprofloxacin for 2 weeks, but patient felt unwell. Then, a combination of moxifloxacin and clarithromycin for 3 months Florid mixed inflammation within the superficial dermis, focally granulomatous in nature. No organisms seen
Clarithromycin for 6 months
Treatment
Granulomatous inflammation with a neutrophilic infiltrate within the dermis. No organisms seen
of granulomatous inflammation identified. No organisms seen Granulomatous inflammation with a neutrophilic infiltrate within the dermis. No organisms seen
Histology
Marked improvement following 3 months of clarithromycin. A further 3-month course was recommended, but the patient failed to attend for follow-up appointment Patient prescribed clarithromycin by GP over a 4-month period and reported improvement on this to his GP. He failed to return to dermatology and has not been contactable by telephone Patient took clarithromycin for 3 months, but continued to have areas of induration and papules within the grey area of the tattoo. Was switched to ciprofloxacin, but it was stopped after 2 weeks owing to side-effects. Then had a 3-month course of moxifloxacin and clarithromyin.
with a good cosmetic outcome
Response to treatment and follow-up
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 85
British Journal of Dermatology (2014) 171, pp79–89
British Journal of Dermatology (2014) 171, pp79–89 Immunosuppressed
No
PMHx
Nil
Age (sex)
28 (F)
Nil
Medication
Histology
Foci of suppurating granulomatous inflammation in superficial dermis. Tattoo pigment was identified. No mycobacterial organisms seen
Culture and sensitivities
Tissue culture positive. Tobramycin, amikacin, ciprofloxacin, linezolid, moxifloxacin, clarithromycin
Distribution and presentation
Arm, tattoo site. Developed papular erythematous eruption within grey-shaded areas of new tattoo
3-month course of clarithromycin
Treatment
Repeat biopsy was negative on culture Patient prescribed 3month course of clarithromycin and then failed to attend the dermatology review appointment. Patient contacted by telephone and stated that the tattoo had improved following 2 months of antibiotics
Response to treatment and follow-up
Number of cases
PMHx, patient medical history; M, male; F, female; PASd, periodic acid–Schiff with diastase; CLL, chronic lymphocytic leukaemia; IV, intravenous; INR, international normalized ratio; GP, general practitioner.
12
Case
Table 2 (continued)
86 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
4 Number of cases in NHS Lothian 2003 –2012
3
2
1
0
Year
Fig 2. Incidence of cases of cutaneous Mycobacterium chelonae infection in the NHS Lothian Health Board area between 1 January 2003 and 31 December 2012.
(a)
(b)
Fig 3. (a, b) Patients with Mycobacterium chelonae infection typically presented with papules and pustules, indurated erythema, erosions and ulcers.
tattooed patients, all inflammatory eruptions were confined to areas of grey shading (Fig. 4). Histologically, a suppurative granulomatous dermatitis/panniculitis was the predominant
© 2014 British Association of Dermatologists
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 87
(a)
(b)
Fig 4. (a, b) All inflammatory eruptions in tattooed patients were confined to areas of grey shading.
pattern of inflammation, occurring in seven of the 12 patients. A purely tuberculoid granulomatous pattern was found in three patients. The remaining two patients showed an abscesslike acute inflammatory picture with granulation tissue-like small vessel proliferation. AFB were detected in four patients (all immunosuppressed), all of whom presented with suppurative granulomatous inflammation with pseudocyst-like empty spaces. None of the tattoo-related cases showed any demonstrable microorganisms on special histochemical stains.
Discussion Mycobacterium chelonae is a ubiquitous environmental organism, and our experience suggests that cases of cutaneous infection appear to be increasing. Previously identified sources of M. chelonae include tap water, water tanks and interventional medical equipment, such as catheters and bronchoscopes. In recent years, infection has also been reported as a complication of cosmetic procedures, including injection of botulinum toxin,3 contaminated foot spas4 and liposuction.5 Outbreaks have also been seen following mesotherapy6 and laser eye surgery.7 The largely unregulated nature of the U.K.’s cosmetic industry and the increasing demand for invasive procedures by the general public may be contributing to the apparent increase in incidence. It is well recognized that immunosuppressed individuals are at particular risk of infections, including cutaneous mycobacterial infections. In our series of 12 patients, all of the six immunosuppressed individuals were on oral steroids, and three were on additional systemic immunosuppression, including biological therapy, cyclophosphamide and leflunomide. None of the immunosuppressed individuals were known to have visceral involvement, but there were two deaths from concur© 2014 British Association of Dermatologists
rent medical problems. We have also reported what we believe to be the first case of M. chelonae at the site of clobetasol propionate steroid application. Of additional interest is our documentation of four confirmed cases of M. chelonae secondary to tattooing, and an additional five further suspected cases in which the clinical features, epidemiological factors and histology were highly suggestive of infection, but skin culture was negative, possibly owing, in part, to concurrent antibiotic treatment. To date, these clusters of patients from Edinburgh and the Lothians are the only cases to be reported in the U.K., but other case series from France and the U.S.A. have been published.8–11 All patients in our series with tattoo-related infection had undergone tattooing with grey ink, and all had very similar eruptions comprising red papules, pustules and induration confined to areas of grey shading. The Civic Government (Scotland) Act 1982 (Licensing of Skin Piercing and Tattooing) Order of 2006 legislates the legal requirements for tattoo parlours in Scotland.12 The Act requires tattoo artists to use only sterile, single-use disposable needles for skin piercing and tattooing, and to use only single pigment or ink prepackaged in single-use vials. However, there is no requirement for tattoo artists to use sterile water for rinsing needles, or diluting down black ink to grey, and it is possible that tattoo artists and ink manufacturers may use tap water for this process. In the first reported cluster of patients in Edinburgh, M. chelonae was cultured from an opened bottle of ink and from rinsings of the ink bottle nozzle, but not from unopened bottles, which was suggestive of environmental contamination. The tattoo parlour implicated in the second outbreak in West Lothian admitted to using nonsterile tap water to dilute black ink down to a grey wash, and also used tap water in a rinse pot used for cleaning needles. Authors of other published cases of tattoo-related M. chelonae infection have also postulated that the source of the infection may be the use of nonsterile water to dilute black ink down to grey, either at a manufacturing level or within tattoo parlours. Consequent to the discovery of tattoo-related M. chelonae clusters in south-east Scotland, a communication was issued to all general practitioners (GPs), consultant dermatologists, microbiologists, plastic surgeons and pharmacists working for NHS Lothian Health Board to highlight the problem and to encourage the referral of possible cases. Public and environmental health departments have been in close contact with the relevant tattoo parlours locally. The outbreaks noted in south-east Scotland and in other countries suggest the necessity of establishing national and international guidelines on sterile techniques in tattoo parlours, including the avoidance of nonsterile water, and tightening of the regulations governing the manufacture and supply of tattoo inks. The cases of cutaneous M. chelonae that we report are similar clinically to other reported cases of cutaneous, nontuberculous mycobacterial infections, including those related to M. marinum, Mycobacterium abscessus and Mycobacterium fortuitum. A case series of 51 patients with cutaneous, nontuberculous mycobacterial infection reported that the most commonly observed lesions British Journal of Dermatology (2014) 171, pp79–89
88 Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al.
were papules and pustules; other patients developed abscesses, inflammatory plaques, nodules and suppurative ulceration.13 All of these clinical features were seen in our patients with M. chelonae infection. These clinical findings, particularly in the context of an immunosuppressed individual, should alert the clinician to the possibility of mycobacterial infection. It is noted that no M. abscessus was isolated from skin specimens across Scotland; the reason for this is unclear. Mycobacterium abscessus is said to be a common cause of skin mycobacterial infection;14 however, reports of this infection from environmental sources are scant, and may represent identification methods that discriminate poorly between M. chelonae and M. abscessus. Alternatively, there may be variable geographical distribution and host interactions accounting for the lack of cutaneous M. abscessus infection in Scotland. In contrast to the apparently increasing incidence of nontuberculous mycobacterial cutaneous infections, cutaneous tuberculosis (TB) remains relatively rare within Europe, although the prevalence differs around the world.15 It is postulated that the incidence within Europe will rise with increased immigration from outside the continent. Cutaneous TB appears to differ clinically from nontuberculous mycobacterial infections, and can take a number of recognized patterns, including lupus vulgaris and scrofuloderma (typically the most commonly reported forms in adults), TB verrucosa cutis, miliary TB and tuberculid. The histological spectrum of nontuberculous mycobacterial infection is rather broad and can be similar to M. tuberculosis infection.16 Cutaneous M. chelonae infections (like the other rapid-growing mycobacterial species, e.g. M. fortuitum) tend to have a distinctive appearance characterized by suppurative granulomatous inflammation surrounding pseudocyst-like structures often containing abundant AFB.17 We have demonstrated a striking difference in the incidence of M. chelonae throughout Scotland. It is possible that there is a higher prevalence of M. chelonae in soil and water in south-east Scotland, but it is, perhaps, more likely that, compared with the rest of the country, there is an increased local clinical and laboratory awareness of the possibility of atypical mycobacteria, resulting in appropriate samples being rapidly submitted and cultured optimally at correct temperatures, with an increased diagnostic pick-up. Most Scottish laboratories culture mycobacteria in liquid or solid media only, and do not culture at both 30 °C and 37 °C, relying on the referral of specimens to the SMRL for specialist culture. The tattoo outbreaks are likely not unique to south-east Scotland, and the lack of confirmed cases in other areas is probably multifactorial, such as patients not seeking medical attention for inflamed tattoos, fewer GP referrals of patients with tattoo reactions, empirical antibiotic treatment clouding the clinical picture, misdiagnosis as allergic contact dermatitis, sarcoid or other granulomatous reaction, or a low index of suspicion for mycobacterial infections in the diagnostic differential, resulting in inadequate or inappropriate investigations. Given the risk factors for developing M. chelonae, including immunosuppression, cosmetic procedures and tattooing, it is possible the incidence will further British Journal of Dermatology (2014) 171, pp79–89
increase in the U.K. Therefore, familiarity with the clinical features, diagnosis and management strategies of cutaneous mycobacterial infection is essential for both dermatologists and nondermatologists to allow prompt diagnosis and initiation of therapy. Management may be guided by antibiotic sensitivities and in consultation with microbiology colleagues. Patients are likely to require a prolonged course of antibiotic therapy, usually for a minimum of 3 months. Clarithromycin monotherapy may often suffice, but some patients may require combinations of antibiotics to ensure resolution. A reduction of immunosuppression may need to be considered in some patients. In particular, the diagnosis should considered in patients with eruptions who are taking long-term systemic glucocorticoids, as our series appears to suggest that they are at high risk of infection.
Acknowledgments We thank colleagues from the Scottish Mycobacteria Reference Laboratory (SMRL) for their assistance. Identification and susceptibility testing is the remit of the SMRL, which is funded through Health Protection Scotland, by National Services Scotland.
References 1 Sergeant A, Conaglen P, Laurenson IF et al. Mycobacterium chelonae infection: a complication of tattooing. Clin Exp Dermatol 2013; 38:140–2. 2 National Records of Scotland (NRS). Census results. Available at: http://www.scotlandscensus.gov.uk/en/censusresults/ (last accessed 24 March 2014). 3 Latorre-Gonzalez G, Garcia-Garcia M, Martinez-Colubi M et al. [Disseminated cutaneous infection by Mycobacterium chelonae after botulinic toxin in an immunosuppressed patient]. Med Clin (Barc) 2005; 125:439 (in Spanish). 4 Sniezek P, Graham B, Busch H, Lederman E. Rapidly growing mycobacterial infections after pedicures. Arch Dermatol 2003; 139:629–34. 5 Meyers H, Brown-Elliott BA, Moore D et al. An outbreak of Mycobacterium chelonae infection following liposuction. Clin Infect Dis 2002; 34:1500–7. 6 Munayco C, Grijalva C, Culqui D et al. Outbreak of persistent cutaneous abscesses due to Mycobacterium chelonae after mesotherapy sessions, Lima, Peru. Rev Saude Pu blica 2008; 42:146–9. 7 Freitas D, Alvarenga L, Sampaoio J et al. An outbreak of Mycobacterium chelonae infection after LASIK. Ophthalmology 2003; 110:276–85. 8 Kluger N, Muller C, Gral N. Atypical mycobacteria infection following tattooing: a review of an outbreak in 8 patients in a French tattoo parlour. Arch Dermatol 2008; 144:941–2. 9 Goldman J, Caron F, de Quatrebarbes J et al. Infections from tattooing. Outbreak of Mycobacterium chelonae in France. BMJ 2010; 341: c5483. 10 Drage L, Ecker P, Orenstein R et al. An outbreak of Mycobacterium chelonae infections in tattoos. J Am Acad Dermatol 2010; 62:501–6. 11 Kennedy B, Bedard B, Younge M et al. Outbreak of Mycobacterium chelonae infection associated with tattoo ink. N Engl J Med 2012; 367:1020–4. 12 legislation.gov.uk. The Civic Government (Scotland) Act 1982 (Licensing of Skin Piercing and Tattooing) Order 2006. Available © 2014 British Association of Dermatologists
Cutaneous M. chelonae in Edinburgh, V.E. Scott-Lang et al. 89 at: http://www.legislation.gov.uk/ssi/2006/43/contents/made (last accessed 24 March 2014). 13 Bartralot R, Garcıa-Patos V, Sitjas D et al. Clinical patterns of cutaneous nontuberculous mycobacterial infections. Br J Dermatol 2005; 152:727–34. 14 Griffith D, Aksamit T, Brown-Elliott B et al. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007; 175:367–416.
© 2014 British Association of Dermatologists
15 Marcoval J, Alcaide F. Evolution of cutaneous tuberculosis over the past 30 years in a tertiary referral hospital on the European Mediterranean coast. Clin Exp Dermatol 2012; 38:131–6. 16 Santa CD, Strayer DS. The histologic spectrum of the cutaneous mycobacteriosis. Hum Pathol 1982; 13:485–95. 17 Kriegsman M, Scott-Lang VE, Holme SA, Biswas A. Suppurative and granulomatous dermatitis with pseudocysts: a useful tissue reaction pattern. Diagn Histopathol (Oxf) 2012; 18:185–8.
British Journal of Dermatology (2014) 171, pp79–89
