Retrospective case series analysing the clinical data and treatment options of patients with a tubercular abscess of the chest wall

Abstract

The tubercular abscess of the chest wall remains one of the differential diagnoses of a chest wall tumour, and the management strategy is controversial. We reviewed the medical records of 22 patients treated at our institution. Two patients were managed by antitubercular medications alone; eight patients were managed by medication and open drainage. Five patients underwent open drainage with subsequent radical surgery at a constant interval of time, and the mean duration between open drainage and radical surgery was 9.8 weeks (range, 3–12). Seven patients underwent radical surgery without prior open drainage. Five patients required rib resections, and curettage of infected pleural peel was necessary in 5 patients. Antitubercular drugs were administered basically for more than 6 months regardless of surgical management, including for more than 1 month prior to radical surgery. Postoperative empyema was seen in 1 patient after radical surgery. The mean follow-up duration was 32.8 months (range, 3–100), and there was no recurrence. Complete resection of the tubercular abscess with sufficient antitubercular therapy resulted in a satisfactory outcome. Antitubercular therapy with or without open drainage can be a viable choice.

INTRODUCTION

Due to the development and widespread use of antitubercular medication, tuberculosis manifesting as a surgical disease is not as common as it once was. Tubercular abscess of the chest wall has been reported to compose less than 10% of skeletal tuberculosis cases [1, 2]. However, when surgeons encounter a mass lesion of the chest wall, it should remain in the differential diagnoses. The diagnosis can be difficult since a smear or culture of acid-fast bacilli (AFB) is often negative. The role of surgery has been changing, and the extent of resection is still controversial [1–3]. The purpose of this retrospective study is to evaluate the clinical characteristics, treatment and outcomes in 22 patients with tubercular abscesses of the chest wall treated in our institution, and to consider the optimal management of this disease.

PATIENTS AND METHODS

We reviewed all the medical records of the patients treated in the ward for tubercular patients of Nishi-Kobe Medical Center in the 15 years between January 1996 and December 2010. We diagnosed 22 patients with tubercular abscess of the chest wall in that period. We confirmed the diagnosis by either a smear of AFB, a culture of the abscess, polymerase chain reaction (PCR) for Mycobacterium tuberculosis, or a biopsy showing epithelioid granulomas with caseous necrosis. The goal of the treatment is the complete removal of abscess by the surgery or the disappearance of abscess by conservative therapy, which meant antitubercular medication with or without open drainage. We reviewed 22 patients to investigate the clinical features, the treatments and the outcome. Our institutional ethics committee approved this retrospective study.

RESULTS

Patient characteristics

The patients' characteristics are shown in Table 1. There were 12 men and 10 women. The mean age of the patients was 63.9 years. The chest wall abscess was left-sided in 12 patients, and right-sided in 10 patients with the average size being 6.4 cm. Seven patients had past histories of tuberculosis, while 8 patients had active tuberculosis concurrently with their chest wall lesion. Seven patients were given a diagnosis of this disease without any history of tuberculosis nor concurrent active disease.

Table 1:

Patients' characteristics

Case	Age/sex	Size (cm)/side	Concomitant Tbc	Past history of Tbc	Underlying disease
1	53/M	8/Rt	Bil. lung
2	43/M	7/Lt	Bil. lung, pharinx, vertebra
3	86/F	5/Rt			Cerebral infarction
4	68/M	6/Rt		R. lung	Rheumatoid arthritis
5	68/M	3/Rt		L. lung
6	54/M	5/Rt
7	81/M	4/Lt		L. lung	Subdural hematoma
8	72/M	5/Rt	Bil. lung
9	95/M	6/Lt			Parkinson's disease
10	29/F	1/Lt	L. lung, MLN
11	62/M	5/Lt		R. lung
12	24/F	7/Lt		R. lung
13	81/F	5/Lt
14	71/M	10/Lt	L. lung		Diabetes mellitus
15	65/F	11/Lt			Urethral carcinoma
16	81/F	18/Rt	L. lung
17	77/M	4/Lt		L. pleura	Thymoma, PRCA
18	73/M	7/Lt			Myocardial infarction
19	85/F	5/Rt		R. lung	Diabetes mellitus
20	69/F	4/Lt
21	26/F	7/Rt	R. lung, R. pleura
22	42/F	7/Rt	R. lung, pharinx, abdominal wall

Tbc, tuberculosis; M, male; F, female; L, left; R, right; Bil, bilateral; MLN, mediastinal lymph node; PRCA, pure red cell aplasia.

Diagnosis

The results of microbiologic evaluation obtained by the aspiration or the drainage of the abscess were as follows: The AFB smear was positive in 12 patients (54.5%), and the culture was positive in 13 patients (59.1%). PCR was performed in 19 patients, and was positive in 16 patients (84.2%). Four patients were diagnosed via histology from the surgical specimen. Computed tomography (CT) scan was done in all cases, and enhanced CT scan was performed in 19 patients. It was not done in 3 patients due to medical problems. A low density central area with rim enhancement (Fig. 1) was recognized in all of those 19 patients. The destruction of adjacent ribs was found in 7 patients (31.8%) and thickening or calcification of the pleura (Fig. 2) on the affected side was seen in 17 patients (77.3%).

Figure 1:

The CT finding showing fluid retention and peripheral rim-enhancement with destruction of the rib (arrow).

Figure 2:

The CT finding demonstrating pleural calcification on the same side of the chest wall abscess (arrows).

Treatments

Antitubercular therapy was started immediately after the diagnosis. Open drainage was performed when the diagnosis was not obtained by an aspiration or the abscess had open fistula and purulent discharge at presentation. In some patients, open drainage had already been done in the other hospital before coming to our institution. We basically recommended radical surgery after preoperative antitubercular therapy in all the patients except for high-risk patients. When an abscess became smaller and smaller through the conservative therapy, we decided whether radical surgery was to be performed, considering the patient's opinion and risk for surgery. The treatment in each patient is shown in Table 2. Two patients (Cases 1–2) were treated by medication only. Eight patients (Cases 3–10) were managed by medication and open drainage of their abscess. These patients refused radical surgery or had some risk factor for surgery such as deconditioning or other comorbidity. Five patients (Cases 11–15) were treated by medication and open drainage, following radical surgery. These patients failed treatment by medication and open drainage. They still had productive, open fistulae at the time of radical surgery. The mean duration between open drainage and radical surgery was 9.8 weeks. Seven patients (Cases 16–22) were treated by medication and radical surgery without open drainage. In the patients undergoing radical surgery, en block resection of the abscess was done in 5 patients. Adequate debridement of the abscess was performed in 7 patients. In 5 patients, involvement of the ribs was recognized and rib resection was performed. The surgical wound was primarily closed in all the patients except for Case 14 undergoing radical surgery. In Case 14, open-window thoracostomy was performed first because large empyema space existed. Decortication and closure of the wound was done 1 month after the surgery. Two patients (Cases 11 and 13) requiring rib resection did not show any change of the ribs in CT scan but did show rib involvement at the time of surgery. The curettage of the plural peel was undertaken in 5 patients for local empyema or intrathoracic invasion of the abscess. The partial resection of the affected lung was done in Case 21 secondary to abscess invasion into the lung tissue.

Table 2:

Treatment and CT findings

Case	Regimen and total length (month)	Duration of drainage (week)	Radical surgery	Rib resection	Rib destructiona	Pleural abnormalitya
1	H, R, E, Z (7)
2	H, R, E, Z (10)				+	+
3	H, R, E, Sm (12)	9				+
4	H, R, E, Z (3)	4				+
5	H, R, E, Sm, Lv (12)	4				+
6	H, R, E, Z (12)	9				+
7	H, R, E (9)	9			+	+
8	H, R, E, Z, Sm, Lv (23)	15				+
9	H, R, E (7)	4			+	+
10	H, R, Sm (15)	1
11	H, R, E, Z, Sm, Lv (15)	10	D	12th		+
12	H, R, E, Z, Sm (6)	12	D
13	H, R, E, Z, Sm (15)	12	D + C	4th, 6th		+
14	H, R, E, Sm (12)	3	D + C	7th, 8th	+	+
15	H, R, E (6)	12	D		+
16	H, R, E (8)		R	6th	+
17	R, E, Z, Sm (12)		R + C			+
18	H, R, Z, Lv, Km (9)		R			+
19	H, R, E, Sm (6)		R + C			+
20	H, R, E, Z (7)		D	10th	+	+
21	H, R, E, Lv, Sm (7)		R + C			+
22	H, R, E, Z, Sm (12)		D			+

H, isoniazid; R, rifampicin; E, ethambutol; Z, pyrazinamide; Lv, levofloxacin; Sm, streptomycin; Km, kanamycin; D, debridement; R, en block resection; C, curettage of pleural peel.

^aCT findings.

In all the patients undergoing radical surgery, antitubercular drugs were administered for more than 1 month before the radical surgery. The average length of total antitubercular medication was 10.2 months.

Outcomes

Cases 1 and 2 showed disappearance of the abscess about 4 months after initiation of antitubercular therapy. In the patients treated by medication and open drainage only, the mean duration required from open drainage to closure of the wounds was 6.9 weeks. Case 4 died 3 months after the initiation of treatment because of acute respiratory distress syndrome. Regarding one postoperative complication, a postoperative empyema occurred in Case 16 and chest tube drainage was performed. The mean follow-up duration was 32.8 months (range, 3–100), and there was no recurrence.

DISCUSSION

Tubercular abscesses located in the chest wall are rare, being less than 10% of skeletal tubercular lesions [1, 2]. Patients usually present with an enlarging, and occasionally painful mass on the chest wall [3]. It is sometimes mistaken as breast tumour in females [4]. A review of the literature suggests the following three mechanisms concerning the aetiology of this disease: direct invasion from adjacent pleural or pulmonary lesions of tuberculosis, haematogenous dissemination, or local extension of lymphadenitis of the chest wall [5, 6]. About 70–80% of the patients with this manifestation have a past history of tuberculosis, and 20–60% of the patients have active tuberculosis [5–7]. It is important to note, however, that patients without any history of tuberculosis nor concurrent active disease can have new chest wall lesions, shown in our series.

Furthermore, diagnosis can be difficult. An aspiration of the abscess has been considered to be usually non-diagnostic. The diagnosis was reported to be confirmed by AFB smear or culture in only 20% of the cases [2, 5, 6]. However, using PCR in addition to a smear or culture, has a higher sensitivity [8]. Our study also supports that the key to making a diagnosis is to perform PCR. The common CT finding of tubercular abscess is a central low-density area with peripheral rim-enhancement [9, 10]. A thickening or calcification of the pleura, and the disruption of adjacent rib is sometimes observed. A CT scan is usually helpful for showing the extent of the abscess, but cannot always determine whether the adjacent rib is involved with the abscess [1, 11]. In our series, 2 of 5 patients did not show destruction of the rib on CT scan, but ultimately required resection of the rib.

The role of surgery and the extent of resection are still controversial. Previously, a long duration regimen of multi-drug antitubercular therapy was considered to be sufficient, and surgery was indicated if the lesion did not improve after 3 months of medication [5, 11–13]. In our series, about 45% of the patients actually improved through medication with or without open drainage, with radical surgery ultimately necessary if the abscess or fistula was still present after about 9 weeks following the start of treatment. It is difficult to determine whether a patient can be successfully treated without radical surgery. Neither the size of the abscess nor the CT finding nor comorbidities seem to correlate entirely with the clinical course.

Paik et al. [14] documented a recurrence rate of 16% after excision only of the abscess, a rate which was decreased to 4.7% after a complete resection of chest wall mass including any involved ribs. Kim et al. [1] also concluded that complete resection with or without skeletal resection could achieve a low recurrence rate (9.2%). Recurrence was usually due to insufficient resection of the involved rib or infected pleura [15]. Concerning the extent of rib resection, we think that partial resection of the involved rib is sufficient and that complete rib removal is not required. In addition, curettage of infected pleural peel should be performed if necessary. Intraoperative evaluation of the extent of the abscess is important since preoperative imaging does not always show the involvement of the rib or pleura. Regarding the risk of radical surgery, operative morbidity was reported to be about 5% [1]. Our complication rates were similar.

Preoperative and postoperative drug therapy is mandatory to achieve acceptable cure rates [6, 7]. If the diagnosis of this disease is confirmed after radical surgery, a sufficient postoperative antitubercular therapy should be given. The duration of antituberculous therapy has been controversial, but a 6- to 9-month regimen including isoniazid and refampicin is considered standard [1, 7, 15]. Thus, complete resection with sufficient antitubercular therapy is considered to offer a satisfactory cure rate with low recurrence, according to the literature and our experience.

Our study has several limitations: its retrospective nature, the small number of patients, and the limited follow-up period. The patients treated by conservative therapy did not show any recurrence in our series, but a longer follow-up and a larger number of patients is necessary to draw any definitive conclusion. This disease can recur more than 5 years after the treatment [1].

We report a satisfactory outcome of the complete resection with adequate antitubercular medication coverage. In high risk patients for surgery, the conservative therapy can be an acceptable management strategy. However, radical surgery sometimes seems unavoidable.

Conflict of interest: none declared.