ARTHRITIS & RHEUMATISM Vol. 39, No. 10, October 1996, pp 1754-1760 6 1996, American College of Rheurnatology 1754

CLINICAL FEATURES AND THERAPEUTIC MANAGEMENT OF SUBGLOTI’IC STENOSIS IN PATIENTS WITH WEGENER’S GRANULOMATOSIS

CAROL A. LANGFORD, MICHAEL C. SNELLER, CLAIRE W. HALLAHAN,

ANTHONY S. FAUCI, and ROBERT S. LEBOVICS GARY S. HOFFMAN, WILLIAM A. KAMMERER, CHERYL TALAR-WILLIAMS,

Objective. To determine the clinical features and optimal treatment of subglottic stenosis (SGS) in pa- tients with Wegener’s granulomatosis (WG).

Methods. Review of 43 patients with SGS and treatment of 20 patients with intratracheal dilation- glucocorticoid injection therapy.

Results. SGS developed in 43 of 189 patients with WG who were followed up at the National Institutes of Health Clinical Center. The diagnosis of SGS occurred in the absence of other features of active WG in 21 of 43 patients (49%). In 21 patients (49%), SGS began while the patient was receiving systemic immunosuppressive therapy for disease activity involving other sites. Tra- cheostomy was required in 10 of 18 patients (56%) who were treated with systemic immunosuppressive therapy. In 20 patients treated with intratracheal therapy, none required tracheostomy and 6 with previous tracheosto- mies were decannulated.

Conclusion. SGS often occurs independently of other features of active WG and is frequently unrespon- sive to systemic immunosuppressive therapy. Intratra- cheal dilation-injection therapy provides a safe and effective treatment for WG-associated SGS and, in the absence of major organ disease activity, should be used

Carol A. Langford, MD, MHS, Michael C. Sneller, MD, Claire W. Hallahan;MS, Gary S. Hoffman, MD (current address: Cleveland Clinic Foundation, Cleveland, Ohio), Cheryl Talar- Williams, MPH, PA-C, Anthony S. Fauci, MD: National Institute of Allergy and Infectious Diseases, National Institutes of Health, Be- thesda, Maryland; William A. Kammerer, MD: Georgetown Univer- sity Medical Center, Washington, DC; Robert S. Lebovics, MD: National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Bethesda, Maryland.

Address reprint requests to Carol A. Langford, MD, MHS, Laboratory of Immunoregulation, National Institute of Allergy and Infectious Disease, National Institutes of Health, Bethesda, MD 20892.

Submitted for publication February 13, 1996; accepted in revised form May 15, 1996.

without concomitant systemic immunosuppressive agents.

Wegener’s granulomatosis (WG) is a necrotizing, granulomatous vasculitis that has a predilection for involving the upper and lower respiratory tracts and kidneys. Stenosis of the subglottic larynx (SGS) due to granulomatous inflammation and scarring occurs in -16% of WG patients and can be potentially life threatening (1-4). Tracheostomy for airway protection has been required in up to 52% of patients, with the reported rate of permanent decannulation varying from 0 to 100% (1,5-7). Although both surgical and medical treatments have been utilized, the optimal therapeutic approach to SGS in patients with WG has not been determined (1,3,5,6). Glucocorticoids in combination with a cytotoxic agent have markedly decreased the mortality and morbidity rates from active WG that affects other major organ systems, but it is not known how effective these agents are in treating the subglottic lesion. Furthermore, such regimens have significant potential toxicities, including infections, and in the case of cyclophosphamide, the risk of hemorrhagic cystitis, bladder carcinoma, myelodysplasia, and permanent ste- rility (2).

Since 1972, 43 patients with WG and SGS have been followed up at the National Institutes of Health (NIH) Clinical Center. During this time, we observed that SGS often occurred independently of other features of disease activity and was not uniformly responsive to systemic therapy. On this basis, we sought to develop other treatments that would optimize the delivery of antiinflammatory medication to the subglottis but would limit patients’ exposure to systemic agents. Since 1991, we have utilized a surgical technique for the manage- ment of SGS which combines mechanical subglottic

SUBGLOTI'IC STENOSIS IN WG 1755

dilation with intratracheal injection of a long-acting glucocorticoid into the stenotic lesion. This report de- scribes the clinical features in 43 individuals with SGS, the results of treating 20 patients with this intratracheal dilation-injection technique, and our current approach to the diagnosis and management of WG-related SGS.

PATIENTS AND METHODS

Patients. Between 1972 and March 1995, 189 patients with WG were followed up for 6 months to 22 years at the Warren Grant Magnuson Clinical Center of the NIH. All patients had biopsy-proven WG with necrotizing vasculitis, granulomatous inflammation, or both, from an involved organ site. Forty-three of the 189 patients (23%) had evidence of SGS, diagnosed either by flexible fiberoptic examination or by direct laryngoscopy. Twenty-five of these patients have been described in an earlier report (1).

Therapeutic approach. Since March 1991, we have used a technique of intratracheal dilation-injection therapy (described below) to treat patients who have significant SGS. Significant SGS was defined as the presence of symptomatic airway compromise, usually consisting of dyspnea, voice changes, or cough, and objective tracheal narrowing, with a tracheal diameter of 5 5 mm for males and 1 4 mm for females. In using this procedure, we have adopted an approach to the treatment of WG-related SGS in which patients who had disease activity isolated to the subglottis were not treated by initiating or increasing systemic immunosuppressive therapy, but instead, underwent local intratracheal dilation-injection therapy. Patients with SGS and active WG affecting a major organ other than the subglottis received systemic immunosup- pressive therapy with glucocorticoids and a cytotoxic agent, as has been previously described (8,9), in addition to intratracheal dilation-injection therapy. Major organ disease activity was defined by the presence of 21 of the following features in the absence of other etiologies: 1) new or progressive lower airways disease; 2) vision-threatening ocular or retroorbital disease; 3) renal disease, as determined by positive findings on renal biopsy or active urinary sediment including red blood cell casts; or 4) new or progressive central or peripheral nervous system disease.

Systemic immunosuppressive therapy was therefore begun, increased, or tapered only as indicated to treat extra- subglottic disease. Twenty patients were treated using this approach, with a median followup time of 35.5 months (range 6-46 months).

Method of intratracheal therapy. Intratracheal dilation-injectiotl therapy was performed concurrently with direct laryngoscopy in the operating room, with the patient under general anesthesia and with jet ventilation. After general anesthesia was achieved, the trachea was suspended using an operative laryngoscope, the trachea and subglottic larynx were examined, and the overall appearance and tracheal diameter were noted. Frequently, a biopsy was performed if the area had an inflamed, ulcerated, or atypical appearance. The intratra- cheal dilation-injection procedure was performed in the same way regardless of whether the trachea had an inflamed or scarred appearance.

Figure 1. Injection of a long-acting glucocorticoid preparation into the stenotic subglottic lesion via an intratracheal approach. The hand of the anesthesiologist is controlling the pressure-regulated jet injector apparatus, which administers ventilation through a side port of the operative laryngoscope.

Tracheal dilation was performed first, using graduated mercury-filled dilators coated with 1 % triamcinolone cream. Following dilation, the stenatic area was injected via an intratracheal approach with methylprednisolone acetate, a long-acting glucocorticoid, in a 4-quadrant submucosal pattern (Figure 1). Approximately half of the glucocorticoid was observed to leak out during the procedure; thus, the syringe was filled with 40 mg of methylprednisolone acetate to yield a total tissue injection dose of 15-25 mg. Immediately before and for 24 hours following surgery, the patient received an intravenous antibiotic with a spectrum that would cover oro- pharyngeal flora. To lessen tracheal swelling related to the mechanical trauma of the procedure, intravenous dexametha- sone (Decadron, 12 mg; Merck, Rahway, NJ) was given during the procedure and hydrocortisone (100 mg) was given imme- diately following the procedure and at 6 hours and 12 hours postoperatively. In patients with normal renal function, intra- venous ketorolac 30 mg every 6 hours was given for up to 24 hours following surgery to lessen discomfort. The patient was observed in the hospital overnight and discharged the follow- ing day if there were no complications.

The trachea was evaluated by the otolaryngologist via indirect or fiberoptic laryngoscopy 5-4 weeks following the procedure. If significant SGS was still present based on these findings together with the patient's symptoms, a repeat intra- tracheal dilation-injection procedure was performed. This procedure would thus be repeated every 3-4 weeks until the patient no longer had significant SGS. Once the patient had no evidence of significant SGS, he or she would be reevaluated by the otolaryngologist at 6-12-week intervals, or sooner if symp- toms returned, and repeat intratracheal procedures were per- formed if significant SGS redeveloped.

RESULTS Clinical features. SGS was a manifestation of

WG in 43 of 189 patients (23%). In 5 patients (2.6%),

1756 LANGFORD ET AL

Table 1. Pattern of organ involvement in 189 patients with Wegen- er’s granulomatosis*

Nb. (%) with No. (%)without subglot t ic subglottic stenosis stenosis (n - 43) (n = i46) P

Sinus Lung Joint Kidney Saddlenose Slun Endobronchial Retroorbital Ocular Nervous system

42 (98) 30 (70) 24 (56) 23 (53) 20 (47) 16 (37) 13 (30) 10 (23) 9 (21.1 8 (19)

119 (82) 125 (86) 95 (65)

34 (23) 112 (77)

(34 (44) 4 (13)

49 (34) 20 (14)

25 (17) ~~

<0.01 0:02 NS

<0.01 <o.oi

NS ;US NS NS NS

* Data were available for only 30 patients in the endobronchial category. NS = not significant (P 2 0.05).

SGS was the presenting feature of WG. Thirty-nine patients (91%) had organ involvement outside of the sinuses and subglottis at some point in their disease course. SGS affected men and women equally, with 27 of 101 females (27%) developing SGS compared with 16 of 88 males (18%) (P = 0.2). The median age at which SGS was diagnosed was 26 (range 13-66). Similar to previous studies, SGS was found to occur more frequently in patients who had the onset of WG at an age <20 (1,lO). Sixteen of 31 patients who had WG onset at <20 years (52%) developed SGS, compared with 27 of 158 patients with WG onset at 220 (17%) (P < 0.001). SGS was diagnosed at a median time of 12 months after the initial onset of WG symptoms (range 0-22 years).

Dyspnea was the most common presenting symp- tom in 34 of the 43 patients with SGS (79%). Voice changes and hoarseness were noted by 26 of the 43 patients (61%), and 10 of the 43 patients (23%) had a cough. Throat discomfort was uncommon and occurred in only 5 patients (12%). An evaluation of other poten- tial risk factors for the development of SGS revealed that 5 patients (12%) had been intubated within 12 months before the diagnosis of SGS, and 1 patient (2%) had extensive herpetic pharyngitis.

In reviewing the overall pattern of organ involve- ment, a greater percentage of the WG patients with SGS had sinus involvement and saddlenose deformity, but a lower percentage had renal and lung involvement, com- pared with the 146 patients with WG who did not have SGS (Table 1). The remaining organ systems were affected at a similar frequency for the 2 populations.

At the time of the onset of SGS, 22 patients (51%) had evidence of active disease affecting other major organ sites, while the remaining 21 patients (49%)

had no evidence of disease activity outside the subglottis. Of these 21 patients, 6 (28.5%) had a major organ system flare within 6 months following the diagnosis of SGS, 6 (28.5%) experienced remission for 212 months before a major organ flare, and 9 (43%) did not have a major organ flare following the diagnosis of SGS. In 21 of the patients (49%), SGS began while he or she was receiving systemic immunosuppressive therapy for dis- ease activity involving other sites (9 taking cyclo- phosphamide plus glucocorticoid, 1 taking methotrexate plus glucocorticoid, 6 taking glucocorticoid, 4 taking cyclophosphamide, and 1 taking azathioprine).

Eleven of the 43 SGS patients (26%) required no surgical therapy for management of their SGS. Of these 11,5 (45%) had no change in medical therapy following the diagnosis of SGS, while the remaining 6 required an increase in immunosuppressive therapy to treat disease activity involving other organ sites. Thirty-two patients (74%) required surgical management; many of them received more than one type of procedure. In addition to the 20 patients who were treated with intratracheal dilation-injection therapy, 17 patients received me- chanical dilation, 9 patients underwent laser resection, 2 patients were treated with repeated laryngoscopic tissue removal, 4 patients underwent laryngotracheoplasty (LTP), and 2 patients had laryngotracheoplasty with microvascular repair (LTP-MVR). Tracheostomy was performed in 18 patients (42%), with 17 later achieving permanent decannulation. The remaining patient re- quired a permanent tracheostomy despite treatment with systemic immunosuppressive therapy, laser and surgical debridement, and tracheal reconstruction. This patient was lost to followup prior to 1991 when we began using intratracheal therapy. Of the 18 patients requiring a tracheostomy, 10 (56%) had received systemic immu- nosuppressive therapy for >2 months prior to undergo- ing tracheostomy (2 taking cyclophosphamide plus glu- coeorticoid, 2 taking cyclophosphamide, 5 taking glucocorticoid, and 1 taking azathioprine).

Thirty-seven of the 43 patients with SGS were being actively followed up at the NIII after May 1990, when testing for the presence of antineutrophil cyto- plasmic antibodies (ANCA) began to be routinely per- formed. Of these 37 patients, 35 were ANCA positive at some point in their disease course, with 30 having a cytoplasmic pattern (cANCA) and 5 having a perinu- clear pattern (pANCA). The 4 patients who had disease localized to the subglottis and sinuses all had a positive ANCA: 3 had cANCA and 1 had pANCA.

Of the 43 SGS patients, 33 had pulmonary func- tion testing with flow-volume (FN) loops performed. Thirteen patients had normal F/V loops; the remaining

SUBGLOTTIC STENOSIS IN WG 1757

1 A

16 ;i./stc J

12; I 1

1

I - 1 ‘ 1 ..

Figure 2. Comparison of flow-volumc loops in A, a normal paticnt and B, a patient with a 2 mm-diameter subglottic stenosis. I n subglottic stenosis. flattening of the inspiratory phase of the loop is seen, which is diagnostic of an extrathoracic airway obstruction.

20 were abnormal. Abnormality was defined as flatten- ing of the inspiratory phase of the loop. In SGS, flattcning occurs in both the inspiratory and expiratory portions of the CUNC, oftcn giving it a box-likc appcar- ance (Figure 2). Howcver, it is the inspiratory flattcning which is most important in assessing SGS. because this is specifically diagnostic of an cxtrathoracic airway ob- struction. Of the 18 patients who required tracheostomy, 10 had F/V loops measured during their course; all of thcm wcre abnormal. Of the 11 patients who had SGS but did not require surgical intcrvcntion. 10 had F/V loops measured. Eight wcrc normal and 2 wcrc abnormal.

Similar to what has bccn prcviously reported, thc diagnostic yield of subglottic biopsies was low (11). A total of 140 subglottic biopsics on 26 patients werc obtained, of which only 7 (5%) revealed diagnostic changes with both granulomatous inflammation plus vasculitis. Acutc and chronic inflammation alone was the prcdominant finding in 93 biopsy samples (66%). Infor- mation was available on the subglottic appearance in 21 1 separate observations of 26 patients made by the same otolaryngologist. The trachea lookcd actively inflamed on 42 occasions and had a scarrcd appcarancc without inflammation at thc remaining 169 examinations. In 31 biopsics obtained from 13 patients when the trachea appeared inflamed, only 1 revealed granulomatous in- flammation plus vasculitis. In comparison, of thc 76 biopsies obtained from 18 paticnts whose trachea did not appear inflamcd, 4 rcvcalcd granulomatous inflam- mation plus vasculitis.

Experience with intratracheal therapy. Of the 43 SGS patients, 20 wcrc trcatcd with intratracheal

dilation-injection therapy under the therapeutic guide- lines outlined abovc. This population consisted of 14 women and 6 men, who had a median age of 31.5 years (range 17-66). Of the 20 patients, 13 required immuno- supprcssivc treatmcnt for concurrent or previously ac- tivc major organ discasc (6 taking methotrexate plus glucocorticoid, 4 taking cyclophosphamide plus glu- cocorticoid, 3, taking glucocorticoid alone, and 1 taking glucocorticoid plus cyclophosphamide followed by glu- cocorticoid plus methotrexate for rclapsc). In 3 of thcsc 13 patients, these agents wcrc tapered and discontinued while the dilation-injection therapy was continued. The remaining 7 paticnts had no evidence of major organ disease activity at that time and wcrc treated with intratracheal dilation-injection therapy alonc. Two of these 7 patients had a pattern of organ involvement that was limitcd to thc subglottis and sinuses, while the remaining patients had previous multisystem disease that was not active at the time of their intratracheal therapy.

The median number of dilation-injection treat- ments each patient received was 3. with a range of 1-21. Six patients requircd only 1 trcatment; the remainder had multiple treatmcnts. In those patients who received multiple treatments, the median length of time betwecn trcatments was 3.7 months, with a range of 1-22 months. The amount of time between procedures was shorter at the beginning of intratracheal treatment and longer thereafter. Within the first 6 months, the median time between procedures was 1.5 months, which increased to a median of 6.6 months following that time. The intra- tracheal dilation-injection procedure was extremcly well tolerated. In 113 procedures performed between March 1991 and March 1995, only 1 complication oc- curred, that being a pneumothorax, which briefly re- quircd insertion of a chest tubc.

The response to intratracheal dilation-injection therapy was assesscd by objcctivc improvement in tra- cheal diameter (Figurc 3) and the need for other surgical interventions or trachcostomy. Ten patients (50%) had rcccivcd surgical treatments other than tracheostomy prior to beginning intratracheal thcrapy. This included 9 patients treated with mcchanical dilation alone, 4 pa- tients trcatcd with laser, 2 patients treated with LTP. and 1 patient treated with LTP-MVR. Following the institution of intratrachcal therapy, no other surgical procedurcs were required on any of the 20 patients in the treated population.

Ninc of the 20 patients had rcquircd a tracheos- tomy prior to beginning dilation-injection treatments. Three paticnts had tracheostomics with permanent de- cannulation prior to beginning dilation-injection thcr-

1758 LANGFORD ET AL

DISCUSSION

WG is a granulomatous vasculitis that most fre- quently presents with upper and/or lower airway symp- toms. SGS has been previously reported to occur in about 16% of WG patients and was seen in 23% of our population of 189 patients. SGS can be a life-threatening manifestation of disease, and prompt diagnosis is essen- tial. A high index of suspicion is important, since the presenting symptoms of SGS can be nonspecific. SGS should be considered in the differential diagnosis of any patient with WG who has increasing dyspnea, voice changes, or cough. Although audible stridor is a helpful diagnostic sign, it is not always present. SGS is optimally diagnosed via tracheal visualization performed by an otolaryngologist. Indirect or fiberoptic laryngoscopy are noninvasive and accurate tracheal examination tech- niques that can be performed in the office by a trained otolaryngologist. Given the life-threatening potential of SGS, all patients in whom SGS is being considered should be evaluated by an otolaryngologist immediately. The presence of SGS may be suggested by flattening of the inspiratory curve on flow-volume loop measurement, which is diagnostic of an extrathoracic airway obstruc- tion (4,7). However, this may not detect less-severe SGS, and should never be used as a primary means of diagnosis. Similarly, magnetic resonance imaging, tra- cheal tomography, or computed tomography should not be used as a primary means of diagnosis because they are insensitive means of viewing the lesion and may lead to an unacceptable delay in diagnosis.

Once a diagnosis of SGS has been established, patients must be monitored closely for signs of airway compromise. Obstruction can occur not only from the subglottic lesion itself, but also from crusted and thick- ened secretions that can result from mucosal inflamma- tion or intermittent upper or lower respiratory tract infections. In patients with known SGS, worsening symp- toms of dyspnea, voice changes, or cough should be immediately evaluated by an otolaryngologist. Such pa- tients may benefit from followup by a single otolaryn- gologist who can become familiar with the appearance of their airway over time. F/V loops and radiographic studies should not be used as a means of monitoring the severity of tracheal stenosis because they are not suffi- ciently sensitive for assessing changes in airway diameter.

The diagnosis of SGS was not found to be reflective of the patient’s overall disease activity. Al- though the diagnosis of SGS should prompt a careful evaluation for evidence of major organ inflammation, the onset of SGS occurred independently of other

Figure 3. Improvement of subglottic diameter from 3 mm (top) to 6 mm (bottom) 1 month after an intratracheal dilation-injection treatment.

apy. The remaining 6 tracheostomies were placed prior to beginning intratracheal treatment and were decannu- lated during treatment. Four of the 6 patients were receiving systemic therapy at the time of decannulation, with the remaining 2 being treated with intratracheal therapy alone. Of the 6 tracheostomies, 3 were repeat procedures, 2 of these repeats were required despite having received systemic immunosuppression. None of the 20 patients required a tracheostomy after beginning intratracheal dilation-injection therapy.

SUBGLOTTIC STENOSIS IN WG 1759

features of active WG in 49% of patients in the study. In these instances, the onset of SGS did not appear to be a harbinger of a disease flare, since only 28.5% experi- enced a major organ flare within 6 months and the remaining 71.5% experienced a flare after 12 months or not at all.

The rationale for avoiding systemic immunosup- pressive therapy in the treatment of isolated SGS is supported not only by the frequent lack of association of SGS with other features of disease activity, but also by evidence that the subglottic lesion is not universally responsive to systemic agents. SGS was diagnosed in 49% of our patients while they were receiving systemic treatment, and 56% of the patients who required trache- ostomies did so despite having been treated for at least 2 months with systemic immunosuppressive agents.

A variety of surgical techniques have been used to treat SGS of other etiologies (12). The success of these in WG-related SGS has been variable, and several of the patients in the present series failed other surgical modalities before achieving benefit from intratracheal dilation-injection therapy. Although they are not re- garded as first-line therapy, intralesional glucocorticoid injections have been successfully used to treat SGS of other etiologies (13-16). This technique was attractive in treating WG-related SGS for several reasons. Intrale- sional glucocorticoid injections have been shown to diminish inflammation and to impair both fibroblast production of collagen and scar formation, possibly through inhibition of cytokine-mediated fibroblast stim- ulation. Glucocorticoids may also bring about collagen resorption and atrophy, as has been noted in keloid and intraarticular therapy, respectively. The addition of me- chanical dilation is important because it disrupts scar tissue that may be present in these lesions and because it immediately widens the patient’s airway. These com- bined mechanisms of action explain not only the means by which this technique may be an effective treatment, but also its applicability in treating all types of WG- related subglottic lesions regardless of whether they are inflamed or scarred.

The decision about when to perform intratra- cheal dilation-injection therapy should be based on the presence of symptoms together with objective evidence of tracheal narrowing. Other parameters such as active systemic disease or a change in ANCA titer should never be used as the basis for considering intratracheal treat- ment. The definition we have used for significant SGS (see Patients and Methods) is intended as a guideline in deciding when to perform intratracheal dilation- injection therapy, but should always be viewed in the setting of the individual patient.

In our hands, the intratracheal dilation- injection technique appeared to be an effective means of treating SGS. In the 20 patients treated using this technique, none have required additional surgical pro- cedures, none have required a tracheostomy, and 6 patients who had tracheostomies at the time treatment was begun were permanently decannulated. The proce- dure was also found to be safe, with only 1 complication occurring in 113 procedures.

In summary, the results of this study indicate that intratracheal dilation-injection therapy is an important technique in the management of WG-associated SGS. Institution of a systemic immunosuppressive regimen with glucocorticoids plus a cytotoxic agent is necessary for the treatment of active WG that is threatening major organs outside of the subglottis. When SGS is present in this setting, intratracheal dilation-injection therapy should be performed concomitantly with systemic ther- apy because of the potential unresponsiveness of the subglottic lesion to these agents. In the absence of other features of active WG, isolated SGS is optimally man- aged with intratracheal therapy alone and does not require the institution of systemic immunosuppressive therapy.

ACKNOWLEDGMENTS

The authors wish to thank the clinical associates, physician assistants, and Clinical Center nursing staff at the National Institute of Allergy and Infectious Diseases for the care of our patients.

REFERENCES

1. Lebovics RS, Hoffman GS, Leavitt RY, Kerr GS, Travis WD, Kammerer W, Hallahan C, Rottem M, Fauci AS: The manage- ment of subglottic stenosis in patients with Wegener’s granuloma- tosis. Laryngoscope 102:1341-1345, 1992

2. Hoffman GS, Kerr GS, Leavitt RY, Hallahan CW, Lebovics RS, Travis WD, Rottem M, Fauci AS: Wegener’s granulomatosis: an analysis of 158 patients. Ann Intern Med 116:488-498, 1992

3. Waxman J, Bose WJ: Laryngeal manifestations of Wegener’s granulomatosis: case reports and review of the literature. J Rheu- matol 13:408-411, 1986

4. Daum TE, Specks U, Colby TV, Edell ES, Brutinel MW, Prakash UBS, DeRemee RA: Tracheobronchial involvement in Wegener’s granulomatosis. Am J Respir Crit Care Med 151:522-526, 1995

5. Arauz JC, Fonseca R: Wegener’s granulomatosis appearing ini- tially in the trachea. Ann Otol Rhinol Laryngol 91:593-596, 1982

6. McCaffrey TV: Management of subglottic stenosis in the adult. Ann Otol Rhinol Laryngol 100:90-94, 1991

7. McDonald TJ, Nee1 HB, DeRemee RA: Wegener’s granulomato- sis of the subglottis and the upper portion of the trachea. A n n Otol Rhinol Laryngol 91588-592, 1982

8. Fauci AS, Haynes BF, Katz P, Wolff SM: Wegener’s granuloma- tosis: prospective clinical and therapeutic experience with 85 patients for 21 years. Ann Intern Med 98:76-85, 1983

9. Sneller MC, Hoffman GS, Talar-Williams C, Kerr GS, Hallahan

1760 LANGFORD ET AL

CW, Fauci AS: An analysis of forty-two Wegener’s granulomatosis patients treated with methotrexate and prednisone. Arthritis Rheum 38:608-613, 1995

10. Rottem M, Fauci AS, Hallahan CW, Kerr GS, Lebovics R, Leavitt RY, Hoffman GS: Wegener granulomatosis in children and ado- lescents: clinical presentation and outcome. J Pediatr 122:26-31, 1993

11. Devaney KO, Travis WD, Hoffman GS, Leavitt RY, Lebovics R, Fauci AS: Interpretation of head and neck biopsies in Wegener’s granulomatosis: a pathological study of 126 biopsies in 70 patients. Am J Surg Pathol 14:555-564, 1990

12. Wannamaker JR, Eliachar I: An overview of treatment options for lower airway obstruction. Otolaryngol Clin North Am 28:7S1-770, 1995

13. Cobb WB, Sudderth J F Intralesional steroids in laryngeal steno- sis. Arch Otolaryngol 96:S2-56, 1972

14. Rosen G, Vered IY: Triamcinolone acetonide injection for laryn- geal stenosis. J Laryngol Otol 89:1043-1046, 1975

15. Waggoner LG, Belensky WM, Clark CE: Treatment of acquired subglottic stenosis. Ann Otol 82:822-826, 1973

16. Gnanapragasam A: Intralesional steroids in conservative manage- ment of subglottic stenosis of the larynx. Int Surg 64:63-67, 1979

Clinical Images: Value of magnetic resonance imaging in extensive pyomyositis

The patient, a 32-year-old woman with a 16-year history of systemic lupus erythematosus (SLE), developed insidious pain and swelling of her right thigh. Her SLE had been quiescent for 3 years with prednisone (7.5 mg/day) and hydroxychloroquine treatment. Ten days after onset, the local symptoms rapidly worsened, as did her general status; fever (39°C) developed. Immediate magnetic resonance imaging (MRI) (1.5 Tesla) revealed typical features of abscessed pyomyositis. Furthermore, the MRI results were helpful to the surgeon in that they delineated the extent of the abscesses, their relationship to the vessels, and the location of associated necrotizing soft tissues. Needle aspiration of the mass yielded a purulent fluid which grew Stuphylococcus aureus, and immediate surgical drainage yielded 1.5 liters of pus. After 4 weeks of antibiotic therapy, the pyomyositis was completely resolved. A, T1-weighted image, transverse view (repetition time [TR] 616 ms, echo time [TE] 15 ms), showing a multilocular mass in the quadriceps muscle with intermediate signal intensity, surrounded by a rim of slight increased signal intensity (arrows). B, T2-weighted image, transverse view (TR 2,280 ms, TE 90 ms), in which the multilocular mass is clearly visible with diffuse high signal intensity, surrounded by muscle tissues with lower but increased signal intensity around the involved muscle (arrowhead). C, Contrast-enhanced T1-weighted image, sagittal view (TR 616 ms, TE 15 ms), showing the presence of a mass extended to two-thirds of the thigh, consisting of multiple lesions with low signal intensity and some areas of heterogeneous signal intensity.

Pascal Claudepierre, MD Bernadette Saint-Marcoux, MD Bruno Larget-Piet, MD Xavier Chevalier, MD, PhD JkrBme Allain Jean-Luc Montazel, MD Hbpital Henii Mondor Criteil, France