The optimal treatment for nontuberculous mycobacterial (NTM) cervicofacial lymphadenitis (CFL) has not been established. Investigators from the Netherlands report the results of a randomized trial of antibiotic treatment compared with surgical excision for NTM CFL. Between September 2001 and December 2004, 100 immunocompetent children with more than three weeks of culture- and/or PCR-documented NTM CFL were randomized to either receive clarithromycin and rifabutin or undergo surgical excision. A single oromaxillofacial surgeon performed all the excisions. Those patients randomized to antibiotic treatment received ≥12 weeks of clarithromycin (15 mg/kg in 2 divided doses) and rifabutin (5 mg/kg once daily).
Patients were followed up at 2, 4, 6, 12, and 24 weeks after start of therapy. Patients judged to have failed antibiotic therapy at six months also underwent surgical excision. Primary outcome was either success or failure of therapy, and secondary outcomes included complications of surgery or antibiotics. Of the 210 children referred, 135 were diagnosed with NTM infection, 100 agreed to participate in the study, and 50 were randomized to each group. The baseline characteristics of the two groups were similar and the majority (82%) of patients was in the late stage of disease with fluctuation of the lymph nodes and skin discoloration. The diagnosis was confirmed by PCR alone in 32, culture alone in 14, and by both PCR and culture in 54.
Cure was defined as regression of lymph node enlargement by ≥75%, with cure of the fistula and total skin closure without local recurrence or de novo lesions after six months, as assessed by clinical and ultrasound evaluation. Surgery was more effective in achieving cure than antibiotic therapy (96% vs 66%; 95% CI for difference, 16%–44%). The NTM infections in 16 patients were resistant to at least one of the two antibiotics, and of those, three were resistant to both but had a favorable response to antibiotic therapy. Stage of disease did not affect outcome. Complications attributed to treatment were seen in 14 patients in the surgery group (including facial nerve palsy or weakness, postoperative infection, and postoperative hematoma) compared with 37 patients in the antibiotic group (including fever, fatigue, abdominal pain, tooth discoloration, headache, vomiting, abnormal stools, and allergic rash). Only one patient in the surgery group had permanent nerve damage and most adverse effects to antibiotics were mild. Two patients in the surgery group required antibiotic therapy with resulting cure compared with 12 patients in the antibiotic group who required surgery for treatment failure.
The authors conclude that surgical excision of NTM CFL clearly results in higher short-term cure rates and recommend that antibiotic treatment be reserved for cases in which surgical excision carries a high risk of facial nerve injury and for those cases involving extranodal infection.
Dr. Dubik has disclosed no financial relationship relevant to this commentary....