J Am Board Fam Pract: first published as 10.3122/jabfm.12.2.158 on 1 March 1999. Downloaded from

Actinomycosis-An Unusual Case of an Uncommon Disease

Louis R. Petrone, MD, jocelyn j. Sivalingam, MD, and Alexander R. Vaccaro, MD

Actinomycosis is a chronic granulomatous disease acetaminophen for the fevers and a triphasic oral caused by any of several anaerobic organisms from contraceptive. She was allergic to codeine. She the genus Actinomyces. Though previously thought smoked one pack of cigarettes per day intermit­ to be a fungal infection, these organisms are actu­ tently for 20 years and drank alcohol minimally. ally caused by gram-positive, filamentous bacteria. She was not currently sexually active; in the past Human infection with actinomycotic organ­ she had consistently used condoms. She denied il­ isms has been recognized since the end of the 19th licit drug use. There was no history of gynecologic century.1-3 In the preantibiotic era, such infections problems, use of an intrauterine device (IUD), or were severe and often fatal. Currently the true in­ bowel disease. cidence of actinomycosis is difficult to define be­ When examined, she was an anxious-appearing cause many mild cases likely are eradicated by the woman whose temperature was 99.5°F, pulse rate common use of antibiotics.2,3 Improved dental hy­ 100 beats per minute, and blood pressure 92/68 giene has also contributed to the declining inci­ mmHg. Findings of examination of head, eyes, dence of infection.2,3 It is generally agreed that ears, nose, and throat were unremarkable; there actinomycosis has become an uncommon entity. was no cervical lymphadenopathy. Her chest was We describe a case of pelvic and sacral actino­ clear, and her heart sounds were regular and with­ mycosis in a patient who had none of the tradi­ out murmurs. There were no abdominal masses, tional risk factors for this disease. tenderness, or hepatosplenomegaly, and there was no edema of the extremities. Tenderness was pre­ Case Report sent over the left sciatic notch. Strength and sen­ A previously healthy 37-year-old woman related a sation in the lower extremities were symmetrical, 2-week history of fevers as high as 104°F associ­ as were reflexes; there was, however, a positive http://www.jabfm.org/ ated with chills, night sweats, nausea, diarrhea, straight leg test (tension sign) indicating irritation anorexia, and malaise. She also reported she had of the sciatic nerve. left lower back and leg pain in the sciatic distribu­ A complete blood cell count revealed a white tion for several months. There were no symptoms blood cell count of 12,000/pL with 73 percent of upper respiratory tract infection, headache, dy­ neutrophils and 6 percent band forms, a hemo­

suria, vaginal discharge, irregular menses, weight globin level of 10.7 g/dL, and platelet count of on 26 September 2021 by guest. Protected copyright. loss, tremors, or myalgias. She had had lower ab­ 507,000/pL. Erythrocyte sedimentation rate was dominal pain, for which she was evaluated by her 103 mm/hr. Alkaline phosphatase was 184 U/L, gynecologist, who noted a fullness in her left ad­ alanine aminotransferase 47 U/L, and gamma­ nexa and recommended a pelvic sonogram. The glutamyltransferase 98 U/L; other liver function sonogram showed an ill-defined left adnexal mass, tests were normal. Urine culture, viral hepatitis and magnetic resonance imaging (MRI) was rec­ studies, antinuclear antibodies, tuberculin test (pu­ ommended for further evaluation. rified protein derivative [PPD]), and chest radi­ Her medical history was notable for a cholecys­ ographs were all negative. The thyroid-stimulat­ tectomy in 1994. She was taking ibuprofen and ing hormone level was normal. Plain radiographs of the lumbosacral spine showed no evidence of abnormally advanced degenerative changes, lytic Submitted 24 March 1998. From the Department of Family Medicine (LRP), the De­ foci or bony destruction, or abnormal soft tissue partment of Medicine (US), and the Department of Orthope­ swelling. dic Surgery (ARV), jefferson Medical College, Philadelphia. MRI of the pelvis showed an infiltrative Address reprint requests to Louis R. Petrone, MD, 2305 Fair­ An mount Ave, Philadelphia, PA 19130. process involving the pyriformis, obturator, ilia-

158 JABFP March-April 1999 Vol. 12 No.2 J Am Board Fam Pract: first published as 10.3122/jabfm.12.2.158 on 1 March 1999. Downloaded from cus, and psoas muscles, the SCIatiC nerve and genital tract. They are not considered particularly plexus, and the left side of the sacrum from S 1 to virulent pathogens, but rather opportunistic ones, S5 with diffuse bony infiltration. The mass was because infection occurs usually only after disrup­ described as mostly solid with some small cystic or tion of the mucous membranes. The disease necrotic areas. There was tethering of bowel loops spreads by direct extension into surrounding tis­ within the pelvis. Our diagnostic considerations sues without regard for tissue planes through the included neoplasia, such as lymphoma, sarcoma, formation of sinus tracts that can lead directly to and malignant fibrous histiocytoma, and infection. the skin. The typical sulfur granules can drain Two biopsies guided by computed tomography from these tracts. Microscopically these infections (CT) showed only skeletal muscle and inflamma­ are lobulated microcolonies of the organism. tory cells. Cultures of the specimens were negative Actinomycosis is traditionally divided into for routine organisms as well as acid-fast bacilli three forms: cervicofacial, thoracic, and abdom­ and fungus. inogenital.!,4 The most frequent site of human in­ Open biopsy of the mass was then performed fection is the cervicofacial area, accounting for through an inguinal approach. At the time of about 40 to 50 percent of cases.!-4 Poor dental hy­ surgery the muscle planes were distorted by a dif­ giene or trauma to the mucous membranes of the fuse inflammatory reaction, making their mobi­ oral cavity is usually responsible for cervicofacial lization and separation difficult. Minor fluctuant actinomycosis.s Approximately 15 percent of foci were palpated within the substance of the il­ actinomycosis occurs in the thorax.1,4 Infection iopsoas muscle complex, which when bluntly here can be caused by direct extension from an ex­ probed expressed approximately 10 to 15 cc of pu­ isting infection in the head and neck region or as rulent fluid. Gram-negative rods and gram-vari­ the result of aspiration of the organism from the able filamentous bacteria were isolated at 36 hours throat. Twenty percent of actinomycotic infec­ from cultures obtained in the operating room. tions occur in the abdomen and pelvis.2,4 The patient was treated empirically with imipe­ Abdominal disease usually results from clinical nem-cilastatin (Primaxin) and began to improve or subclinical disruption of bowel mucosa. It often clinically. The organisms were eventually identi­ occurs as a firm mass that appears fixed to the sur­ fied as Fusobacterium nucleatum and Actinomyces is­ rounding tissue and can be mistaken for tumor.2,6-8 raelii. Pathologic section revealed sulfur granules An intra-abdominal event, such as appendicitis or compatible with actinomycosis. perforated viscus, could give rise to pelvic actino­ http://www.jabfm.org/ When the causative organisms were identified, mycosis.9,10 In the early 1970s a rise in the inci­ the patient was given an 8-week course of high­ dence of pelvic infection was attributed to in­ dose intravenous penicillin followed by an ex­ trauterine contraceptive devices (IUDS).l1 When tended course of oral clindamycin. She has re­ pelvic actinomycosis occurs, it usually causes en­ mained afebrile since starting the antibiotics and 2 dometritis, salpingo-oophoritis, or tubo-ovarian

years later feels well and has returned to work full­ abscess,12 and a mass in the adnexa might be palpa­ on 26 September 2021 by guest. Protected copyright. time. Follow-up MRIs with gadolinium have ble, suggesting a pelvic malignancy/,8 Ultimately, shown substantial improvement in the inflamma­ extension to the abdominal wall or deep pelvic tory mass but persistent changes in the sacrum, structures can occur. the importance of which are unclear. Follow-up Actinomycosis can also occur in bone, usually sedimentation rates have been normal. as the result of direct extension from a nearby in­ fection. 13 In the pre antibiotic era, the vertebrae Discussion were frequently infected bones, predictive of a Actinomycosis is a bacterial infection that can af­ mortality rate as high as 76 percent}4 Recently, fect virtually any site in the body. Disease in hu­ however, given the relative frequency of orofacial mans is most commonly caused by A israelii.! actinomycosis and the decrease in the occurrence Other species capable of causing human infection, of widespread disease, the facial bones, particu­ although less frequently, are Actinomyces odontolyti­ larly the mandible, are the most frequent sites of cus and Actinomyces viscosus.2 These organisms have osseous infection.14 Current studies have found been isolated from mucous membranes of the that the incidence of vertebral infection is quite mouth, bronchi, gastrointestinal tract, and female low. In one study of 181 cases, lung abscesses ex-

Actinomycosis 159 J Am Board Fam Pract: first published as 10.3122/jabfm.12.2.158 on 1 March 1999. Downloaded from

tended to vertebral bone in only two cases. 15 esting for several reasons. First, she had no identi­ Other smaller studies found no cases of vertebral fiable source of pelvic infection; that is, she had involvement. 1,4,5 never used an IUD, nor did she have a history of Hematogenous spread to other organs, such as abdominopelvic trauma or disease. She had had an the brain, liver, and kidney, can occur but is quite uncomplicated cholecystectomy approximately 1 uncommon. Central nervous system involvement year before onset of her symptoms, and although occurs in approximately 3 percent of cases, usually actinomycosis of the gallbladder has been re­ as a result of dissemination from a primary lung ported,19 it is unlikely that her gallbladder was the focus 3 or extension of cervicofacial disease. source of infection because there was no evidence Actinomycosis has been noted to be a poly­ of direct extension from the right upper quadrant. microbial infection2-4; it is often found along with She had had no other gastrointestinal or pelvic another organism, such as Fusobacterium in our procedures before her symptom onset. Findings patient. Other bacteria that are isolated with Ac­ from a barium enema and a follow-up gynecologic tinomyces include anaerobic streptococci, certain examination after the diagnosis was made were Haemophilus species, and various gram-negative completely normal. Hematogenous seeding is un­ bacilli. 1 It is possible that these organisms act as likely because other sites of disease would be ex­ cofactors that allow the Actinomyces organisms to pected throughout the body. Furthermore, there propagate. was no history of recent dental work, and findings Diagnosis of actinomycosis can be difficult be­ from a dental examination after the diagnosis was cause of the insidious nature of the infection. made were completely normal. We postulate that Many times the diagnosis is missed preoperatively, a microperforation could have occurred in her even after fine-needle aspiration. 16 The finding of bowel or genital tract that allowed access of the sulfur granules from any site other than the tonsils organisms to the deep pelvis. Another possibility is considered pathognomonic. I-3 CT- or sono­ is that with routine sexual intercourse the organ­ graphic-guided biopsy can be used to obtain mate­ isms could have been expressed through the fal­ rial for diagnosis. Occasionally, as with our pa­ lopian tubes into the pelvis. tient, surgery could be required. Another unusual aspect of this case is that our Once the organism is identified, treatment is patient had extensive sacral involvement of infec­ relatively straightforward. Most strains of Actino­ tion, which to our knowledge has not been re­

myces are sensitive in vitro to a variety of agents, ported. http://www.jabfm.org/ such as tetracycline, erythromycin, cephalo­ Finally, neurologic complications of actinomy­ sporins, and clindamycin. 16 Ciprofloxacin and cosis are usually the result of direct pressure of the imipenem-cilastatin have also been usedy,18 It is inflammatory mass on a nerve or on the spinal generally recommended, however, to treat with cord. Cases of epidural involvement by Actino­ penicillin, to which the organisms are usually sen­ myces organisms resulting in paraparesis have been sitive, because of its low cost, its high tolerability, reported.20,21 Our patient had symptoms of sciatic on 26 September 2021 by guest. Protected copyright. and the extent of clinical experience. 1,5 Long-term nerve compression, a result of compression by the therapy, lasting from 8 weeks to several years, is infectious mass. Sciatica has not been reported in usually recommended. 1,9 Generally, the disease is several series of cases of actinomycosis.4,5,15 treated until there is evidence of complete resolu­ tion. Surgery is occasionally needed to drain ab­ Summary scesses.5,9 but because actinomycotic infection Actinomycosis is an uncommon disease caused by does not follow tissue planes, surgery can be com­ organisms of the Actinomyces genus. These organ­ plicated and, if possible, should be delayed at least isms are commonly found in the mucous mem­ until after a course of antibiotics has been given. branes but do not cause infection unless there is The prognosis of actinomycosis in the current disruption of the membranes, as occurs, for exam­ era of antibiotics is usually very good. If diagnosis ple, during dental trauma or abdominal surgery. is delayed, however, extensive local involvement Use of an IUD is also a risk factor for pelvic actin­ can develop. Furthermore, if the disease dissemi­ omycosis. The disease is usually insidious and is nates, especially to the brain, death can uccur.l often mistaken for other conditions. Treatment of The evolution of disease in our patient is inter- the infection, once diagnosed, is a regimen of

160 }ABFP March-April1999 Vol. 12 No.2 J Am Board Fam Pract: first published as 10.3122/jabfm.12.2.158 on 1 March 1999. Downloaded from long-term antibiotics such as penicillin, clin­ of actinomycosis. South Med J 1986;79: 1574-8. damycin, and others. Our patient had pelvic and 10. Ellis LR, Kenny GM, Nellans RE. Urogenital as­ pects of actinomycosis.J UroI1979;122:132-3. sacral actinomycosis without any of the traditional 11. Henderson SR. Pelvic actinomycosis associate with risk factors for infection. an intrauterine device. Obstet Gynecol 1973;41: 726-32. References 12. ValicentiJF Jr, Pappas AA, Graber CD, WIlliamson 1. Bennhoff DF. Actinomycosis: diagnostic and thera­ HO, WIllis NF. Detection and prevalence oflUD­ peutic considerations and a review of 32 cases. associated Actinomyces colonization and related Laryngoscope 1984;94: 1198-217. morbidity. A prospective study of 69,925 cervical 2. Russo TA. Agents of actinomycosis. In: Mandell smears.JAMA 1982;247:1149-52. GL, BennettJE, Dolin R, editors. Mandel, Douglas 13. Young WB. Actinomycosis with involvement of the and Bennett's principles and practice of infectious vertebral column: case report and review of the liter­ disease. 4th edition. New York: Churchill Living­ ature. Clin Radiol 1960; 11: 175 -82. stone, 1995. 14. Lewis RP, Sutter VL, Finegold SM. Bone infections 3. Actinomycosis. In RipponJW Medical mycology: involving anaerobic bacteria. Medicine-Baltimore the pathogenetic fungi and the pathogenic actino­ 1978;57:279-305. mycetes. 3rd edition. Philadelphia: WB Saunders, 15. Brown JR. Human actinomycosis: A study of 181 1988. subjects. Hum Pathol 1973;4:319-30. 4. Weese WC, Smith 1M. A study of 57 cases of actino­ 16. Fiorino AS. Intrauterine contraceptive device-asso­ mycosis over a 36-year period. A diagnostic 'failure' ciated actinomycotic abscess and Actinomyces detec­ with good prognosis after treatment. Arch Intern tion on cervical smear. Obstet Gynecol 1996;87: Med 1975;135:1562-8. 142-9. 5. Eastridge CE, Prather JR, Hughes FAJr, YoungJM, 17. Macfarlane DJ, Tucker LG, Kemp RJ. Treatment of McCaughan JJ Jr. Actinomycosis: a 24 year experi­ recalcitrant actinomycosis with ciprofloxacin. J In­ ence. South MedJ 1972;65:839-43. fect 1993;27:177-80. 6. Fowler RC, Simpkins KC. Abdominal actinomycosis: 18. Yew WW, Wong PC, Wong CF, Chau CH. Use of a report of three cases. Clin RadioI1983;34:301-7. imipenem in the treatment of thoracic actinomyco­ 7. Perlow JH, WIgton T, Yordan EL, GrahamJ, Wool sis. Clin Infect Dis 1994;19:983-4. N, WIlbanks GD. Disseminated pelvic actinomyco­ 19. Smithers BM, Wall DR, Weedon D. Actinomycosis sis presenting as metastatic carcinoma: association of the gallbladder. Aust N ZJ Surg 1983;53:587-8. with the Progestasert intrauterine device. Rev Infect 20. MuIIer PG. Actinomycosis as a cause of spinal cord Dis 1991;13:1115-9. compression: a case report and review. Paraplegia 8. Hinnie J, Jacques BC, Bell E, Hansell DT, Milroy R. 1989;27:390-3.

Actinomycosis presenting as carcinoma. Postgrad 21. Kannangara DW, Tanaka T, Thadepalli H. Spinal http://www.jabfm.org/ MedJ 1995;71:749-50. epidural abscess due to Actinomyces israelii. Neurol­ 9. Wohlgemuth SD, Gaddy Me. Surgical implications ogy 1981;31:202-3. on 26 September 2021 by guest. Protected copyright.

Actinomycosis 161