Concomitant Interstitial Pneumonia and Disseminated BCG Infection after Intravesical BCG Therapy

Abstract

An 83-year-old man presented with persistent fever after intravesical BCG therapy for bladder cancer. Chest computed tomography (CT) and bronchoscopy revealed diffuse ground-glass opacities with multiple micronodules and lymphocyte-predominant bronchoalveolar lavage fluid with a high CD4/CD8 ratio, respectively. Therefore, corticotherapy for interstitial pneumonia was initiated. Anti-fast staining of the respiratory specimens was negative; however, anti-tuberculosis treatment was added based on CT findings suggesting disseminated BCG infection. Fifty days post-admission, Mycobacterium bovis BCG was identified in the initial sputum cultures. Concomitant interstitial pneumonia and BCG infection should be considered in patients with abnormal chest imaging following intravesical BCG therapy.

Introduction

Intravesical BCG therapy has been widely used as an adjuvant therapy after tumor resection in high-risk non-muscle invasive bladder cancer (NMIBC) (1). BCG consists of an attenuated derivative of Mycobacterium bovis BCG (M. bovis BCG)-a member of the Mycobacterium tuberculosis (M. tuberculosis) complex-which can cause severe side effects in the lungs after intravesical injections, including interstitial pneumonia and disseminated BCG infection (2,3).

Herein, we report a case of bladder cancer complicated by interstitial pneumonia and disseminated BCG infection after intravesical BCG therapy.

Case Report

An 83-year-old man with a 12-pack-years smoking history was diagnosed with NMIBC and had undergone transurethral resection of bladder tumors 15 and 5 months previously. To avoid further recurrence, the patient had been treated with a cycle of intravesical BCG therapy (80 mg at a time, once a week for 6 weeks; Tokyo 172 strain, Japan BCG Laboratory, Tokyo, Japan) 4 months previously. However, 1 week after the first dose of the second cycle of intravesical BCG therapy, the patient experienced fever, loss of appetite, and general fatigue persisting for 4 days. Therefore, the patient was urgently admitted to our hospital (day 1). The patient's medical history included hypertension with regular intake of calcium channel blockers. There was no history of dust or asbestos inhalation or environmental exposure to birds or molds.

Upon admission, the patient's vital signs were as follows: body temperature of 37.9°C, blood pressure of 124/68 mmHg, heart rate of 72 beats/min, and saturation of percutaneous oxygen of 94% on room air. The patient had clear breath sounds, no heart murmur, and no leg edema.

Table presents the results of the blood and urine tests. Notably, the blood count was normal, the C-reactive protein was 4.02 mg/dL, and the KL-6 levels were within normal limits. However, the transaminase levels were elevated (aspartate transaminase, 291 U/mL; alanine transaminase, 298 U/mL), and the patient presented with hematuria and pyuria.

Table.

Laboratory Findings upon Admission.

Hematology
White blood cells	6,000	/μL
Neutrophils	77.8	%
Lymphocytes	17.7	%
Basophils	0.3	%
Eosinophils	0.2	%
Monocytes	4.0	%
Red blood cells	498×104	/μL
Hemoglobin	15.6	g/dL
Platelets	25.3×104	/μL
Biochemistry
TP	6.8	g/dL
Alb	3.2	g/dL
BUN	18.0	mg/dL
Cre	0.89	mg/dL
AST	291	IU/L
ALT	298	IU/L
ALP	232	IU/L
T-bil	2.2	mg/dL
Serology
CRP	4.02	mg/dL
KL-6	435	U/mL
Urinalysis
Red blood cells	10-19	/HPF
White blood cells	10-19	/HPF

Alb: albumin, ALT: alanine transaminase, ALP: alkaline phosphatase, AST: aspartate transaminase, BUN: blood urea nitrogen, Cre: creatinine, CRP: C-reactive protein, HPF: high power field, T-bil: total bilirubin, TP: total protein

Chest radiograph showed bilateral diffuse shadows (Fig. 1A), and high-resolution computed tomography (HRCT) of the chest revealed the new appearance of diffuse ground-glass opacities and multiple micronodules at the apex of the lungs (Fig. 2). Additionally, computed tomography (CT) revealed hepatomegaly. We suspected bacterial pneumonia, interstitial pneumonia, and disseminated BCG infection after BCG therapy as the causes of the persistent fever and abnormal chest shadow. After initiating antibiotic treatment (cefazolin), transbronchial lung biopsy and bronchoalveolar lavage (BAL) were performed on day 3. Although the lung biopsy did not reveal any granulomatous lesions, the BAL fluid (BALF) revealed an increased total cell count (490×10³/mL), hyperlymphocytosis (70%, 343×10³/mL), and an increased CD4/CD8 ratio (7.36). On day 4, the patient received high-dose methylprednisolone (1,000 mg for 3 days) followed by prednisolone [60 mg/day (1 mg/kg)] and anti-tuberculosis therapy, including isoniazid (INH) of 300 mg/day, rifampicin (RFP) of 450 mg/day, and ethambutol (EB) of 750 mg/day (Fig. 3).

Figure 1.

Chest radiographs showed bilateral diffuse shadows on day 4 (A). After corticosteroid and anti-tuberculosis therapies, the shadows resolved on day 10 (B).

Figure 2.

Computed tomography images obtained 7 months prior to admission, i.e., before the first intravesical BCG therapy (A, C, E), and at admission (B, D, F). Multiple micronodules in the apex of the lungs (arrowheads) and diffuse ground-glass opacity in the bilateral lungs appeared upon admission.

Figure 3.

Clinical course of the present case. ALT: alanine transaminase, EB: ethambutol, CRP: C-reactive protein, INH: isoniazid, mPSL: methylprednisolone, PSL: prednisolone, RFP: rifampicin

Although the urine smear and polymerase chain reaction tests were positive for the M. tuberculosis complex group, those tests of the sputum, BALF, and blood were negative. Therefore, disseminated BCG infection could not be determined biologically, and anti-tuberculosis therapy was terminated on day 11 due to the prolonged elevation of liver enzymes. Owing to improvements in clinical symptoms and resolution of bilateral diffuse shadows on chest radiographs (Fig. 1B), corticosteroid therapy was tapered, and the patient was discharged on day 28. However, on day 50, M. bovis BCG was detected in two sputum samples obtained upon admission. Therefore, the corticotherapy was terminated as planned, and anti-tuberculosis therapy with INH, RFP, and EB was administered at the same dose as before to treat the disseminated BCG infection. After the anti-tuberculosis treatment, liver enzymes levels did not increase, and chest radiographs showed no worsening of the granular shadows.

Discussion

The present case was novel, as intravesical BCG therapy resulted in two pulmonary complications. Previously, separate cases of interstitial pneumonia and microbiologically diagnosed disseminated BCG have been reported; however, only a few cases of concomitant complications have been reported (2-4).

Frequent side effects of intravesical BCG include complications of low-grade fever (less than 38.9°C) and malaise within 48 hours of administration, occurring in 28% and 24% of patients, respectively (5), and these are considered to be an immune response to BCG. In contrast, a fever of more than 39.5°C or persistent fever for (72 hours may indicate serious systemic side effects that require careful monitoring (2,3).

Interstitial pneumonia after intravesical BCG infusion therapy is a rare side effect, with an incidence of 0.7% (2). In such cases, chest radiography and CT often reveal bilateral infiltrating shadows and diffuse ground-glass opacity patterns, respectively (2,6). In previous cases, bronchoscopy revealed an increased cell count with lymphocyte dominance and an elevated CD4/CD8 ratio in the BALF (7), whereas non-caseating granulomas were observed upon lung biopsy (2). In the present case, the chest imaging findings and BALF results were consistent with those of previous reports. The elevation of KL-6, which is commonly used for diagnostic support and the assessment of clinical activities in drug-induced interstitial pneumonia (8), is not necessarily observed in this disease (9,10). In general drug-induced interstitial pneumonia, KL-6 levels are increased in specific HRCT patterns with fibrosis (8). In the present case, CT images predominantly showed diffuse ground-glass opacity without architectural distortion, suggesting a hypersensitivity pneumonitis pattern without elevated KL-6 levels.

Since interstitial pneumonia is thought to occur due to a hypersensitivity reaction to the protein component of BCG, steroids are considered an appropriate treatment (7). However, in a previous report of 21 cases in Japan, the mortality was 23.8% even with steroid therapy - including high-dose steroid therapy (11) - indicating that BCG-related interstitial pneumonia has a potentially severe outcome.

In a previous single-center study, disseminated BCG infection-a severe systemic adverse event occurring with lung involvement after intravesical BCG therapy-occurred in 4.3% of patients (3). In this pooled analysis of 282 patients with BCG infection (3), approximately half of the cases were diagnosed microbiologically, while the other half were diagnosed based on symptoms of persistent high fever, abnormal chest imaging consistent with typical miliary patterns or biochemical findings (liver and kidney function), and responses to anti-tuberculosis drug treatment with no other differential disease. Since the rate of positive urine cultures 6 days after BCG intravesical administration is 16% (12), with some cases even having positive urine cultures 3 months after the last administration (13), urine specimens are not reliable for determining whether BCG infection is present during a certain period after administration. Microbiological diagnosis can also be challenging; therefore, chest imaging changes may better assist in determining the presence of BCG infection. Since disseminated BCG infection is a serious infection, with a reported mortality rate of 9.9% (3), early treatment should be considered upon detection. Owing to the pyrazinamide resistance observed among all strains of M. bovis, anti-tuberculosis treatment with INH, RFP, and EB is usually recommended for 9 months after infection (14).

In addition to abnormal chest shadows, this case involved hepatic injury with hepatomegaly. The hepatitis could have been caused by disseminated BCG infection or associated with interstitial pneumonia through a similar hypersensitivity reaction mechanism (15). Although a liver biopsy was not performed to differentiate between these two conditions, gradual reduction in liver enzyme levels following interruption of the initial anti-tuberculosis treatment and continuation of steroid treatment suggested that the hepatitis was due to a hypersensitivity reaction.

Previously, Molina et al. stated that although distinguishing between interstitial pneumonia and disseminated BCG infection with BCG intravesical infusion therapy poses certain difficulties, simultaneous treatment with anti-tuberculosis drugs and steroids should be initiated in a timely manner due to the need for early intervention in cases of suspected disease (16). However, the occurrence of hepatitis caused by disseminated BCG infection or hypersensitivity reaction may interfere with anti-tuberculosis drug therapy since the adverse effect of hepatotoxicity is a concern (17).

In the present case, the imaging findings and bronchoscopy results predominantly indicated interstitial pneumonia and the radiographic and clinical findings improved immediately after initiating steroid therapy. Thus, the main etiology during hospitalization was considered to be interstitial pneumonia. Although chest imaging also suggested disseminated BCG infection, a positive urine antimicrobial test 10 days after the last BCG injection was not sufficient to indicate BCG infection, and it was difficult to definitively diagnose BCG infection until sputum culture results were available on day 50. This delay in the microbiological diagnosis of BCG infection and prolonged hepatotoxicity affected the decision to continue anti-tuberculosis treatment.

Although concomitant interstitial pneumonia and disseminated BCG infection are rare, this case underscores the importance of considering BCG infection with concomitant interstitial pneumonia in patients presenting with related abnormal chest imaging following intravesical BCG therapy-even when the initial microbiological tests are negative-and treating them with particular attention to the side effects.

The authors state that they have no Conflict of Interest (COI).