Kumar, Anand R. MD; Grewal, Navanjun S. MD; Katchikian, Hurig V. BS; Jones, Neil F. MD
UCLA Hand Center, Division of Plastic and Reconstructive Surgery and Department of Orthopedic Surgery, University of California, Los Angeles, CA.
The authors received no funding, grants, or in-kind support for the research or the preparation of the article.
The authors had no association or financial involvement with any organization or commercial entity having a financial interest in or financial conflict with the subject matter or research presented in the article.
aWritten informed consent was obtained from each patient according to the policies of the institutional review board at UCLA Medical Center.
Address correspondence and reprint requests to Navanjun S. Grewal, MD, 733 N. Kings Rd, Apt 221, Los Angeles, CA 90069. E-mail: firstname.lastname@example.org.
Mycobacterium marinum is a slowly growing bacterium that may cause disease in fish and people. This pathogen was initially isolated in 1926, but its link to causing human hand disease was first reported by Linell and Norden in 1951. In the United States, most human infections caused by this bacterium had been associated with exposures to fish tanks. After exposure, symptoms usually appear within 2 to 4weeks. Infections detected early can be treated with antibiotics. However, antibiotics may need to be taken for as long as 18 months or longer. Aggressive M. marinum may cause extensive osteomyelitis and amputation of the involved digit.1 In some instances, surgical drainage and debridement are necessary to preserve hand function. We present 2 cases of M. marinum extensor tenosynovitis that failed to improve with triple antimicrobial therapy and eventually required surgery.
A 33-year-old white female schoolteacher presented to her local physician with dorsal hand and wrist swelling. She had cleaned a very dirty aquarium 3 weeks prior. Her medical history included long-standing type 1 diabetes requiring insulin and psoriatic arthritis requiring methotrexate weekly. She routinely checked her glucose levels with fingertip sticks. After initial treatment with steroid injections, she developed 2 draining sinus tracts that grew M.marinum (antimicrobial susceptibilities, Table 1). Treatment with a 3-drug regimen (rifampicin, ethambutol, and minocycline) was initially successful. She was considered a failure after approximately 12 weeks of oral therapy because her draining nodules recurred. On presentation, she had a discrete boggy swelling overlying the dorsoradial aspect of her left wrist and over the dorsal aspect of her left hand around the extensor tendons of the first, second, third, and fourth compartments. There were 2 draining sinuses over the dorsal aspect of the wrist and over the dorsal aspect of the small finger metacarpal (Fig. 1A). She had full flexion and extension of the fingers and thumb but some restriction of the extremes of wrist flexion and extension. She underwent radical tenosynovectomy of the extensor tendons over the dorsum of the hand and wrist (Fig. 1B) followed by no further antimicrobial therapy. The surgical specimen was recultured and grew M.marinum. In addition, the histopathology was consistent with M. marinum. Her postoperative course was complicated by mild wound breakdown that responded to dressing care. Her infection was eradicated, but she required tenolysis of the extensor tendons and manipulation of the metacarpophalangeal joints 1 year later. Her final range of motion was 60 degrees of wrist flexion and extension and normal flexion and extension of the fingers (Fig. 1C).
A 54-year-old white male schoolteacher presented to his local physician with a chronic draining wound on the dorsum of his middle finger. He reported scraping the dorsum of this finger on a nail and was then exposed to seawater in a museum aquarium. He was initially treated with steroid injections and prolonged empiric levofloxacin treatment but developed satellite lesions on the dorsum of his hand and forearm requiring 2 drainage procedures 5 months and 7 months later. Cultures grew M. marinum (antimicrobial susceptibilities, Table 1). Based on the susceptibility testing, he was then started on antimycobacterial therapy with a 3-drug regimen of rifampicin, ethambutol, and minocycline for 12 weeks. Because of persistent drainage, he was ultimately referred for further evaluation that revealed a healed incision over the dorsal aspect of the middle finger and boggy swelling over the dorsal hand and 2 draining sinuses (Fig. 2A). Preoperative range of wrist motion was 45 degrees of flexion and 65 degrees of extension and nearly normal motion of the thumb and fingers. He underwent radical tenosynovectomy of the extensor tendons of the wrist and dorsal hand (Fig. 2B). The draining sinuses healed uneventfully, and no further masses have appeared on his wrist, hand, or fingers. After surgery, 12 weeks of the same antibiotic regimen (rifampicin, ethambutol, minocycline) were given. Postoperative range of motion was similar to preoperative measurements, and there has been no evidence of recurrence. Cultures remain negative for M. marinum.
More than 60 cases of M. marinum infection of the hand and upper extremity have been reported in a variety of specialty-specific journals. The number of reported cases of nontuberculous mycobacterial infections has steadily increased since 1965. This is thought to be related to the decrease in incidence of tuberculous infections and the increasing awareness of nontuberculous mycobacterial infections. Mycobacterium marinum is a ubiquitous aquatic microbe. Known carriers include freshwater and saltwater fish, mollusks, and snails. All reported cases had confirmed direct exposure to aquatic environments and or fish.2,3 Commonly used nicknames include swimming pool granuloma, fish tank granuloma, fish handlers' nodules, and fish fancier's disease. Many cases have been reported in persons with home aquariums. Exposure while swimming in infected pools, baths, rivers, and reservoirs has also been associated with this infection. Infection can occur from bites and fin trauma but also just from contact with water where preexisting hand lacerations provide a portal of entry for the mycobacterium.
Patients with M. marinum infection typically present with subacute lesions usually on the hands, forearm, or elbows with fish exposure and on the lower extremities with pool exposure. An incubation period of up to 4 weeks precedes the appearance of skin lesions. They appear as single or multiple nodules or papules with associated crusting, erythema, or ulceration.4 Swelling of nearby lymph nodes occurs because spread of infection is usually along lymphatic channels. Rarely though, the axillary lymph nodes are involved. If the infection has progressed and spread to deeper structures, fluctuant areas or draining sinuses may occur. Diagnosis is often delayed because of the rarity of this condition, and it is frequently confused with fungal, bacterial, or parasitic infections.5 Bacteremia and rapidly progressing infections have occasionally been described inthe immunocompromised patient.6,7 If an infection is suspected in such persons, a health care provider should be promptly consulted.
The predilection of M. marinum infections in the extremities is related to the inhibition of microbial growth at temperatures greater than 37°C. The cooler extremities are therefore more predisposed to become infected. Incisional or excisional biopsy with subsequent microscopy and microbiological culture of the specimen is required to confirm the diagnosis. M. marinum grows best on Lowenstein-Jensen media at temperatures of 30°C to 33°C. The typical incubation period is 1 to 3 weeks. It is important to alert the laboratory that the M. marinum species is suspected because incubation at 37°C, which is ideal for M. tuberculosis, will inhibit growth of the atypical species. Either caseating or noncaseating granulomatous inflammation combined with an acute inflammatory reaction is often seen upon histological examination.
Three types of infection have been defined according to the depth of penetration and involved structures.8,9 Type 1 lesions are spontaneously resolving superficial skin lesions. These lesions are usually treated with antimicrobial therapy alone. Surgical excision is reserved for refractory lesions. Type 2 lesions are subcutaneous in nature and may have overlying ulceration. These lesions should be excised primarily followed by antimicrobial therapy. Any residual masses remaining after antimicrobial therapy should be excised. Ill-defined lesions that present during therapy typically resolve, and excision is not necessary.
Type 3 lesions are granulomatous infections of the tendons, joints, or bones. These lesions need to be aggressively resected and/or debrided as necessary. Radical tenosynovectomies, synovectomies, arthrodesis, and even amputation have been reported. Three major series have been reported that specifically address the surgical treatment of atypical mycobacterial infections of the hand or upper extremity, and all support the concept that radical surgical treatment of type 3 lesions is necessary for eradication of infection.
Kozin and Bishop10 reviewed 12 patients infected with M. marinum. Patients had various upper-extremity site lesions that were all culture positive for M. marinum. Seventy-two surgical procedures including debridements, biopsies, and tenosynovectomies were performed, with each patient undergoing an average of 2 procedures. The average duration of antimicrobial drug therapy was 10 months with a single-agent regimen. Infectious disease consultants followed patients. The overall recurrence rate was 24% that may be caused by the inclusion of immunodeficient patients whose overall cure rate was only 36% compared with 87% cure rate for the immunocompetent patients.
Hurst et al11 reviewed 15 patients with M. marinum infections primarily of the hand-21 lesions involved the fingers, and 6 lesions involved the wrist. All lesions were culture positive for M. marinum. Thirty-one surgical procedures were performed, with most patients similarly requiring at least 2 operations for cure. Disease eradication was achieved in all patients, with delayed healing in 2 patients.
Last of all, Chow et al12 reviewed 24 patients with M. marinum type 3 lesions. His group initially used an antibiotic regimen reserving surgical treatment for medical failure. This was the only prospective study of the 3 series that evaluated treatment options for this rare infection. Lesions again primarily involved the hand-19 finger and 5 wrist lesions. Again, all lesions were culture positive for M. marinum. All patients were initially treated with ethambutol and rifampicin for a mean duration of 9 months. Forty-five percent of patients had complete resolution of infection with medical treatment only, and 13% of patients had complete resolution with medical therapy alone but with delayed healing. The remaining 41% of patients required aggressive operative treatment for eradication of their infections. Twenty-one patients (87.5%) had a similar range of motion as the opposite unaffected hand at final follow-up. Prognostic factors for failure of medical treatment alone were history of prolonged pain, draining sinus tracts, and instillation of local steroids into these lesions.
Rifampicin and ethambutol are used most often in invasive infections, although many therapeutic choices exist (antimicrobial susceptibilities, Table 2). In the study by Aubry et al13 on 61 clinical isolates, rifamycins and clarithromycin were the most potent, with the lowest minimum inhibitory concentrations (MICs), and resistance was uncommon. Doxycycline, ethambutol, and minocycline all showed higher MICs but were still effective. Rhomberg and Jones14 tested 11 agents against 37 clinical isolates and found that trimethoprim/sulfamethoxazole was the most potent agent, but 92% of isolates were susceptible. Clarithromycin and minocycline, by contrast, showed susceptibility rates approaching 100% and retained similar potency. This study reported an MIC50 for most quinolones of 4 mg/mL or higher. Braback et al15 found approximately three fourths of isolates susceptible to ciprofloxacin and levofloxacin, and 100% of M. marinum isolates were susceptible to gatifloxacin. Among newer antibiotics tested against M. marinum in this series, only linezolid showed good in vitro activity, but there is limited published clinical experience. The organism is classically resistant to isoniazid and pyrazinamide. Most isolates produce β-lactamase, and in vitro use of a β-lactamase inhibitor and clavulanic acid produces acceptable MICs.16 On the basis of the sparse data correlating susceptibility testing results to clinical response, and the relative infrequency of resistance, recent guidelines suggest foregoing susceptibility testing in M. marinum infections, unless the infection does not respond to treatment.17 Most cases of invasive M. marinum infection require surgical debridement. This approach seems particularly appropriate in immunocompromised patients, those with tenosynovitis, or those for whom medical therapy fails.
Optimal treatment of M. marinum has not yet been established.18 The infection probably resolves spontaneously in some cases, although complete resolution may take up to 2years. At present, no study has compared different antibiotic regimens because of the small number of cases and the difficulty and delay in making a correct diagnosis. In the literature, various antibiotic combinations have been described, but no prospective studies or strong consensus toguide antimicrobial treatment exists. Currently, noncomplicated infection can be treated with fluoroquinolones (ciprofloxacin/levofloxacin), trimethoprim-sulfamethoxazole, doxycycline, or clarithromycin alone or in combination. A 2-drug combination therapy is probably optimal. The preferred oral regimens is clarithromycin 500 mg BID plus ethambutol 15 mg/kg per day or clarithromycin 500 mg BID plus rifampicin 600 mg/d. A rifampicin-containing regimen is preferred by some for osteomyelitis or deep structure involvement. Other active oral agents include minocycline 100 mg BID, doxycycline 100 mg BID, trimethoprim-sulfamethoxazole 1 DS TAB BID (DS = 160 mg trimethoprim/800 mg sulfamethoxazole).
Duration of therapy for superficial infections ranges from 6 to 20 weeks depending on the severity of infection and clinical response. For immunocompetent persons, stopping antibiotics 4 to 6 weeks after clinical resolution may be considered; however, typical minimum antibiotic duration is 3 to 4 months. Longer duration of therapy may be required if infection spread from skin to deeper structures, and in immunocompromised patients. In vitro susceptibilities can be used to guide antimicrobial treatment particularly if a change in therapy is necessary because of poor clinical response. Type 2 and 3 infections are typically treated with combination therapy of ethambutol and rifampicin with duration of therapy ranging from 6 to 18 months. Recurrence of skin lesions may occur in immunocompromised patients after discontinuation of antimicrobial therapy. The goal of early treatment is to hasten time to recovery and prevent progression to deep tissue infection. The natural history of this infection is either spontaneous resolution over 1 to 6 years or progression of infection to deeper tissues.
Mycobacterium marinum infections are an uncommon cause of subacute or chronic infections of the hand and upper extremity. A high index of suspicion is necessary for early diagnosis of M. marinum if a skin nodule or reddened sore (ulcer) develops after direct skin contact with freshwater or saltwater or after handling or processing fish. Early superficial lesions can be treated with antimicrobial therapy alone with good results. Deeper subcutaneous infections or lesions are best excised. Aggressive deep infections involving bone, joints, or tendons are better treated with early radical debridement followed by prolonged antimicrobial therapy.
1. Clark RB, Spector H, Friedman FM, et al. Osteomyelitis and synovitis produced by Mycobacterium marinum in a fisherman. J Clin Microbiol. 1990;28:2570-2572.
2. Zenone T, Boibieux A, Tiguad S, et al. Non-tuberculosis mycobacterial tenosynovitis: a review. Scand J Infect Dis. 1999;31:221-228.
3. Kelly PJ, Karson AG, Weed LA, et al. Infection of synovial tissues by mycobacteria other than Mycobacterium tuberculosis. J Bone Joint Surg Am. 1967;49:1521-1530.
4. Lau JKH, Chow SP, Stroebel AB, et al. Mycobacterial infections of the hand: clinical features and functional results. Hand. 1983;15:192-200.
5. Brown JW III, Sanders CV. Mycobacterium marinum infections. A problem of recognition, not chemotherapy? Arch Intern Med. 1987;147:817-818.
6. Parent LJ, Salam MM, Appelbaum PC, et al. Disseminated Mycobacterium marinum infection and bacteremia in a child with severe combined immunodeficiency. Clin Infect Dis. 1995;21:1325-1327.
7. Hanau LH, Leaf A, Soeiro R, et al. Mycobacterium marinum infection in a patient with the acquired immunodeficiency syndrome. Cutis. 1994;54:103-105.
8. Jones EJ. Mycobacterium marinum hand infections. Br J Plast Surg. 2000;53(2):161-165.
9. Bhatty MA. Mycobacterium marinum hand infections: case reports and review of literature. Br J Plast Surg. 2000;54(2):181.
10. Kozin SH, Bishop AT. Atypical mycobacterium infections of the upper extremity. J Hand Surg. 1994;19A:480-487.
11. Hurst LC, Amadio PC, Badalamente MA, et al. Mycobacterium marinum infections of the hand. J Hand Surg. 1987;12:428-435.
12. Chow SP, Ip FK, Lau JHK, et al. Mycobacterium marinum infections of the hand and wrist. Results of conservative treatment in twenty-four cases. J Bone Joint Surg Am. 1987;69A:1161-1168.
13. Aubry A, Chosidow O, Caumes E, et al. Sixty-three cases of Mycobacterium marinum infection: clinical features, treatment, and antibiotic susceptibility of causative isolates. Arch Intern Med. 2002;162:1746-1752.
14. Rhomberg PR, Jones RN. In vitro activity of 11 antimicrobial agents, including gatifloxacin and GAR936, tested against clinical isolates of Mycobacterium marinum. Diagn Microbiol Infect Dis. 2002;42:145-147.
15. Braback M, Riesbeck K, Forsgren A. Susceptibilities of Mycobacterium marinum to gatifloxacin, gemifloxacin, levofloxacin, linezolid, moxifloxacin, telithromycin, and quinupristin-dalfopristin (Synercid) compared to its susceptibilities to reference macrolides and quinolones. Antimicrob Agents Chemother. 2002;46:1114-1116.
16. Flynn CM, Kelley CM, Barrett MS, et al. Application of the E-test to the antimicrobial susceptibility testing of Mycobacterium marinum clinical isolates. J Clin Microbiol. 1997;35:2083-2086.
17. Woods G. Susceptibility testing for mycobacteria. Clin Infect Dis. 2000;31(8):1209-1215.
18. Donta ST, Smith PW, Levitz RE, et al. Therapy of Mycobacterium marinum infections: use of tetracyclines vs. rifampin. Arch Intern Med. 1986;146:902-904.
© 2008 Lippincott Williams & Wilkins, Inc.