Auris Nasus Larynx 42 (2015) 305–310
Contents lists available at ScienceDirect
Auris Nasus Larynx journal homepage: www.elsevier.com/locate/anl
Laryngeal tuberculosis: A report of 17 cases Masashi Kurokawa a,*, Ken-ichi Nibu b, Kei-ichi Ichimura c, Hiroshi Nishino d a
Department of Otolaryngology, Osaka Prefectural Medical Center for Respiratory and Allergic Diseases, Japan Department of Otolaryngology-Head and Neck Surgery, Kobe University Graduate School of Medicine, Japan c Department of Otolaryngology, Ishibashi General Hospital, Japan d Department of Otolaryngology-Head and Neck Surgery, Jichi Medical University, Japan b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 December 2014 Accepted 7 February 2015 Available online 3 March 2015
Objective: The aim of this study was to reduce misdiagnosis and inappropriate treatment of laryngeal tuberculosis. Methods: Between April 2009 and March 2013, 1660 inpatients with tuberculosis were treated at the Osaka Prefectural Medical Center for Respiratory and Allergic Diseases. Seventeen of these patients were diagnosed with laryngeal tuberculosis. Criteria used for diagnosis were findings of histologic examination of biopsy specimens (2 patients) or clinical response of granuloma to antituberculosis therapy (15 patients). Medical records were used for this retrospective study. Results: Patients’ age ranged from 30 to 84 years with an average of 51.4 14.0 years (mean SE); nine were men and eight were women. The most frequent chief complaint was hoarseness (n = 15). Thirteen patients had persistent cough before they showed otolaryngologic symptoms. Endoscopic findings were categorized into five types of lesions: perichondritic (n = 6), ulcerative (n = 6), granulomatous (n = 6), polypoid (n = 1) and nonspecific inflammatory (n = 1). Laryngeal tuberculosis showed as a single lesion in one patient and as multiple lesions in the other patients. The most commonly involved site was the true vocal cord (n = 16, 94.1%), while 16 patients showed radiographic evidence of active pulmonary tuberculosis. These results indicate that deterioration of the lesions tended to result in the occurrence of multiple lesions. Conclusion: While morbidity in tuberculosis has been decreasing in Japan, a significant number of patients still has laryngeal tuberculosis in association with advanced pulmonary tuberculosis. If a patient has a protracted cough, laryngeal tuberculosis should be taken into consideration for differential diagnosis. Pulmonary imaging study should be performed for early diagnosis and to prevent the infection from spreading. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tuberculosis Larynx Perichondritic lesions Ulcerative lesions Granulomatous lesions Polypoid lesions
1. Introduction In the pre-antibiotic era, laryngeal tuberculosis used to be common and usually was associated with advanced cavitary pulmonary lesions. The incidence of laryngeal tuberculosis with pulmonary tuberculosis was about 40% [1,2]. Many patients with laryngeal tuberculosis died at that time due to severe odynophagia and dyspnea caused by laryngeal edema and granulation. After the development of chemotherapy, improvements in living standards
* Corresponding author at: Department of Otolaryngology, Kishiwada City Hospital, 1001, Gakuhara-Cho, Kisiwada-City, Osaka 596-8501, Japan. Tel.: +81 072 4445 1000; fax: +81 072 441 8812. E-mail address: [email protected] (M. Kurokawa). http://dx.doi.org/10.1016/j.anl.2015.02.012 0385-8146/ß 2015 Elsevier Ireland Ltd. All rights reserved.
and the spread of public health, laryngeal tuberculosis has become quite rare and now occurs in less than 1% of tuberculosis cases [3]. However, tuberculosis still remains a major global health problem. In 2012, an estimated 8.6 million people developed tuberculosis and 1.3 million died from the disease, including 320,000 deaths among the HIV-positive population [4]. In Japan, even though prevalence of tuberculosis has been decreasing every year since 1999, there are still more than 21,000 new patients in each year with a prevalence of tuberculosis of 16.7 per 100,000 populations in 2012 [5]. This figure is still higher than that of most developed countries. Thus, Japan belongs to tuberculosis middle burden country. In addition, AIDS and other immunosuppressive diseases or treatments also have increased and can thus be expected to further increase the incidence of tuberculosis [6,7]. However, laryngeal tuberculosis is often misdiagnosed and it thus takes time to determine the appropriate treatment,
306
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
resulting in the spread of infection, which can become a public health problem [8]. To reduce misdiagnosis and inappropriate treatment of laryngeal tuberculosis, we examined the clinical characteristics of laryngeal tuberculosis. 2. Methods Between April 2009 and March 2013, 1,660 inpatients with tuberculosis were treated at Osaka Prefectural Medical Center for Respiratory and Allergic Diseases. Seventeen of these patients were diagnosed with laryngeal tuberculosis. Criteria used for diagnosis were findings of histologic examination of biopsy specimens (2 patients) or clinical response of granuloma to antituberculosis therapy (15 patients). Patients’ medical records were retrospectively reviewed to obtain data regarding their age, presenting symptoms, site and type of the lesions, and findings of chest X-ray and CT scan, sputum smear, sputum culture, biopsy, polymerase chain reaction (PCR) tests, and QuantiFERON test (QFT). The distinguishing features of laryngeal tuberculosis will be discussed and compared with those reported elsewhere in Section 4. The data were also analyzed for primary diagnosis, patient’s delay (interval between onset of the disease and first visit to a doctor) and doctor’s delay (interval between first visit to a doctor and diagnosis as tuberculosis). The data were analyzed with Dr. SPSS II (SPSS Inc., Chicago, IL, USA), and presented as mean standard deviation (SD). An independent sample t-test was used to determine p-values. Count data were analyzed with Fisher’s exact test and p < 0.05 was considered statistically significant. 3. Results 3.1. Demography and past history The patients’ age ranged from 30 to 84 years with an average of 51.4 14.0 years (mean SE). Nine patients were men with an age range of 30–84 years (average 52.9 15.0 years) and eight were women with an age range of 33–68 years (average 50.0 12.1 years). There were two patients (11.8%) with diabetes (type 1: one, type 2: one), two with mental disorders (schizophrenia: one, social withdrawal: one), and one patient (5.9%) each with bronchial asthma and laryngeal cancer. Nine patients (52.9%) had a history of smoking. 3.2. Endoscopic findings
Fig. 1. Laryngoscopic findings. (A, B) Perichondritic lesions. (A) At start of therapy. (B) 5.5 months later. Erosion in the whole larynx and swelling of left arytenoid and the epiglottis was observed in 66-year-old man. (C, D) Ulcerative lesions. (C) At start of therapy. (D) At improvement. Whitish ulcerative lesion of both true vocal was observed in 44-year-old man. (E, F) Granulomatous lesions. (E) At start of therapy. (F) 4 months later. Granuloma-like swelling in both true vocal cords was observed in 51-year-old man.
Table 1 Findings of endoscopic laryngeal examination by category. Endoscopic findings
The various laryngoscopic findings were categorized into five different types of lesions: perichondritic (n = 3, 17.6%) (Fig. 1A), ulcerative (n = 6, 35.2%) (Fig. 1C), granulomatous (n = 6, 35.2%) (Fig. 1E), polypoid (n = 1, 5.9%), and nonspecific inflammatory (n = 1, 5.9%) (Table 1). Laryngeal tuberculosis manifested as a single lesion in only one patient, and the other 16 patients (94.1%) showed multiple lesions, with nine having severe lesions (perichondritic and ulcerative). 3.3. Symptoms and involved laryngeal subsites The most frequent chief complaint was hoarseness (n = 15, 88.2%), followed by odynophagia (n = 7, 41.2%) and abnormal sensation syndrome of the throat (n = 4, 23.5%). Thirteen patients (76.5%) had persistent cough before manifesting otolaryngologic symptoms. The infected subsites were true vocal cord (n = 16, 94.1%), arytenoids (n = 6, 35.3%), false vocal cord (n = 4, 23.5%) and epiglottis (n = 4, 23.5%). Deterioration of the lesions tended to be resulting in the involvement of multiple lesions (Table 2).
Single
Multiple
Total
%
Perichondritic lesions Ulcerative lesions Granulomatous lesions Polypoid lesions Nonspecific inflammatory lesions
0 0 1 0 0
3 6 5 1 1
3 6 6 1 1
17.6 35.2 35.2 5.9 5.9
Total
1
16
17
Epiglottis
True vocal cord
False vocal cord
Arytenoid
3 0 1 0 0
3 5 6 1 1
3 0 1 0 0
1 4 1 0 0
100
Table 2 Tuberculous sites involved in larynx. Endoscopic findings Perichondritic lesions Ulcerative lesions Granulomatous lesions Polypoid lesions Nonspecific inflammatory lesions
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
3.4. Imaging diagnosis and sputum examination Sixteen patients showed radiographic evidence of active pulmonary tuberculosis (94.1%) and chest radiographic findings were normal for only one patient even on HRCT (high-resolution CT). The sputum smear test was positive for tuberculosis for 16 patients (94.1%). Ten patients had highly contagious infection (58.8%). Results of the sputum culture and polymerase chain reaction (PCR) test were available for 17 patients; all of whom showed positive results (Table 3). 3.5. Pathologic examination and blood test (QFT) Two patients underwent a biopsy; both of whom were pathologically diagnosed with laryngeal tuberculosis. The QuantiFERON1 (QFT) test was used for two younger patients (30 and 33 years old); both of whom were positive for QFT.
307
Table 3 Results of various investigative procedures for 17 cases of laryngeal tuberculosis (PCR: polymerase chain reaction test, QFT: QuantiFERON test).
Positive Negative No record Sensitivity (%)
Chest X-ray
Chest CT
Smear test
Culture
PCR
Histology
QFT
16 1 0 94.1
16 1 0 94.1
16 1 0 94.1
17 0 0 100
13 0 4 100
2 0 15 100
2 0 15 100
Table 4 Diagnoses by otorhinolaryngologists and general practitioners. Otorhinolaryngologist Laryngeal tuberculosis Pulmonary tuberculosis Chorditis Polypoid vocal cord Glottic edema Laryngeal papilloma Common cold
(N = 8)
General practitioner
(N = 9)
2 1 1 1 1 1 1
Pulmonary tuberculosis Bronchitis Laryngeal tumor Mycoplasma pneumonia
4 2 2 1
3.6. Primary diagnosis and patient’s delay, doctor’s delay, and total delay Before coming to our Center, eight patients (47.1%) visited an otorhinolaryngologist (ENT group) and nine patients (52.9%) visited a general practitioner (GP group). The first diagnoses made by these doctors are summarized in Table 4. Patient’s delay (interval between onset of the disease and first visit to a doctor) was 181.9 113.5 days for the ENT group, and 100.6 73.9 days for the GP group. Doctor’s delay (interval between first visit to a doctor and diagnosis of tuberculosis) was 35.4 31.6 days for the ENT group, and 20.3 30.4 days for the GP group. The corresponding values for total delay (interval between onset of the disease and diagnosis of tuberculosis) were 217.3 110.7 and 120.9 60.7 days. Total delay for the ENT group was significantly longer than for the GP group (p < 0.05). Mean patient’s delay was 171.7 137.3 days for the patients with perichondritic lesions, 127.3 85.5 days for those with ulcerative lesions, 160.2 83.9 days for those with granulomatous lesions, 60 days for those with polypoid lesions, and 60 days for those with nonspecific inflammatory lesions. These values for patient’s delay were not significantly different. In relation to endoscopic findings, doctor’s delay was 3 2.8 days for perichondritic type, 20.8 22.4 days for ulcerative type, 41.3 37.5 days for granulomatous type, 54 days for polypoid type and 30 days for nonspecific
inflammatory type. Doctor’s delay for perichondritic lesions was significantly shorter than for granulomatous lesions (p < 0.05). 3.7. Smoking Nine patients (52.9%) had a history of smoking. Patient’s delay was 90.6 72.6 days for smokers and 193.1 93.4 days for nonsmokers, while doctor’s delay was 16.9 27.3 days for the former and 39.3 28.6 days for the latter. Both types of delay were longer for smokers than for nonsmokers. Patient’s delay for smokers was significantly shorter than for nonsmokers (p < 0.05) (Fig. 2). 3.8. Treatment and outcome For patients less than 80 years old, the standard treatment regimen lasted 6 months (2HREZ/4HR:INH/RFP/EB/PZA 2 + INH/ RFP 4) and for patients 80 years old and older, it lasted 9 months (2HRE/7HR:INH/RFP/EB 2 + INH/RFP 7). However, treatment had to be extended for 11 patients due to side effects for four cases, intractable disease for four cases, diabetes mellitus for one case, multiple drug-resistant tuberculosis for one case, and side effects and diabetes mellitus for one case. Ultimately, all patients attained complete remission.
Fig. 2. Difference of days for delay between smokers and nonsmokers.
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
308
Laryngeal tuberculosis also showed a favorable response to the treatment in all cases (Fig. 1A–F). Improvement of symptoms occurred within several weeks from the initiation of medical treatment. Complete resolution of laryngeal lesions was obtained within several months except for two patients with perichondritic type lesions. Of the five patients with simultaneous tracheal tuberculosis, three showed progressive tracheal stenosis in spite of medication. One of them underwent tracheostomy, one underwent tracheoplasty and one remained with complete pulmonary atelectasis. 4. Discussion 4.1. Demography and past history Thaller and Hunter reported tuberculosis is most common in the fourth to sixth decades [9,10], and this accords with the age
range for our study population. The predominance of males has long been a feature of most reports, although the male-to-female ratio varies from 1.5:1 to 9:1 [11,12]. Diabetes has been considered as a significant risk factor as also found in our series. The statistics for tuberculosis in Japan show that in 2012, 14.3% of all tuberculosis patients had diabetes and that figure has been increasing [5]. Two patients in our series had severe mental disorders, and their patient’s delay was 1 year and 9 months, while their doctor’s delay was only 1 day, suggesting that mental disease is a risk factor for the spread of tuberculosis infection. Although there were no patients with HIV among our study population, 13% of patients with tuberculosis were complicated with HIV infection in the world. In particular, approximately 43% of patients with tuberculosis in Africa were complicated with HIV infection in 2013 [4]. In fact, tuberculosis has become a major worldwide issue for patients with HIV.
Fig. 3. Decision tree for laryngeal tuberculosis. Continuous arrow: positive findings; dashed arrow: negative findings.
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
309
4.2. Endoscopic findings
4.6. Treatment outcome
The most common infection subsite was identified as the true vocal cord 40 years ago and this was also true for our series [13– 17]. As a result, the most common chief complaint has been hoarseness for patients of all ages [13–15]. As for type of appearance, about half of our patients had severe laryngeal lesions (perichondritic or ulcerative lesions) and multiple subsites were involved in most of the patients. In the pre-antibiotic era, granulomatous type tuberculosis accounted for only 10% of all laryngeal tuberculosis in the 1940s. During the period 1960–1980, however, granulomatous lesions reportedly accounted for more than 60% [16–18], probably owing to the development of antibiotics and laryngoscopy. However, more than half of the lesions in our series showed severe forms such as perichondritic and ulcerative lesions, and this ratio surpassed that reported by Sasaki et al. [18]. These results suggest that there is a tendency for patients with tuberculosis to consult medical practitioners when they are at a more advanced stage.
Laryngeal tuberculosis is known to show favorable antituberculous response [9,19]. Indeed, in most cases the symptom improve within several weeks and the larynx nearly recovers its original condition within several months. However, surgical treatment, although rarely used, may be necessary if the disease results in airway obstruction or laryngeal fibrosis [19]. We do not biopsy larynx lesions once lung tuberculosis has been confirmed, since a biopsy is not appropriate or necessary in the presence of a large number of bacteria. When a benefit agent is administered for 2 weeks, the infectivity of tuberculosis can be reduced by a factor of 10,000 [25]. Thus, when the response to tuberculosis treatment remains poor during the first several weeks, a laryngeal biopsy for differential diagnosis of laryngeal cancer should be considered.
4.3. Image diagnosis and sputum examination There are two major theories to account for infection associated with laryngeal tuberculosis [14,19]. The first theory is known as the bronchogenic theory and posits that the larynx is infected by the direct spread of large numbers of bacilli from lung tuberculosis. The other theory, known as the hematogenous and lymphogenous theory, states that tubercle bacilli spread via the lymphogenous and hematogenous routes from an early infectious lesion. A study by Auerbach found that 698 (92.3%) of 700 autopsied cases of laryngeal tuberculosis had advanced pulmonary tuberculosis [1]. While it has been reported that primary laryngeal tuberculosis has been increasing in recent years [8,13,20], the majority of cases show positive findings on X-rays in not only our study but also in others [14,19,21]. Those studies support that bronchogenic route is the main cause of infection. 4.4. Initial diagnosis and delay Total delay for the patients in the ENT group was significantly longer than that for those in the GP group. It is a fact that otorhinolaryngologists tend to approach the disease from the viewpoint of laryngeal findings, while general practitioners tend to do so on the basis of imaging-based evidence. The fact that the initial diagnosis made by otorhinolaryngologists varied widely, strongly suggests the importance of chest X-ray for correct diagnosis. Indeed, 16 of the 17 patients in our study had lung tuberculosis that could be detected by regular chest X-ray. 4.5. Smoking In our study, patient’s delay and doctor’s delay for smokers were shorter than those for nonsmokers. Nevertheless, endoscopic findings for smokers showed more progressive stages than those for nonsmokers. While several reports have dealt with the correlation between smoking and lung pulmonary tuberculosis [22], as far as we know no report has been reported on the correlation between smoking and laryngeal tuberculosis. Smoking affects the larynx mucous membrane through chronic stimulation and leads to diminished defensive capability of the larynx [23,24]. These considerations suggest that the laryngeal mucous membrane inflamed by smoking may be more likely to be affected by the transairway infection from tuberculosis and thus to result in more progressive laryngeal tuberculosis.
5. Conclusions While morbidity of tuberculosis has been decreasing in Japan, there are still numerous patients with laryngeal tuberculosis associated with advanced pulmonary tuberculosis. If the patient has a protracted cough in addition to hoarseness, not only macroscopic examination of the larynx, but also an early imaging study too is required as summarized in Fig. 3. Quite a few patients with laryngeal tuberculosis associated with advanced pulmonary tuberculosis seem to consult an otorhinolaryngolist for the initial diagnosis. However, provision of an early pulmonary imaging study and treatment to prevent the infection from spreading are also of vital importance. Conflict of interest None of the authors have any conflict of interest, financial or otherwise. References [1] Auerbach O. Laryngeal tuberculosis. Arch Otolaryngol 1946;44:191–201. [2] Harbersohn SJ. The treatment of laryngeal tuberculosis. J Laryngol 1905;20: 630–7. [3] Willams RG, Douglas-Jones T. Mycobacterium marches back. J Laryngol Otol 1995;109:5–13. [4] World Health Organization (WHO). Global tuberculosis report; 2013. [5] Tuberculosis year book. Tokyo, Japan: Japan Anti-Tuberculosis Association; 2013. [6] Burns JL. Laryngeal tuberculosis. J Otolaryngol 1993;22:398. [7] Yencha MW, Linfesty R, Blackmon A. Laryngeal tuberculosis. Am J Otolaryngol 2000;21:122–6. [8] Kandiloros DC, Nikolopoulos TP, Ferekidis EA, Tsangaroulakis A, Yiotakis JE, Davilis D, et al. Laryngeal tuberculosis at the end of the 20th century. J Laryngol Otol 1997;111:619–21. [9] Thaller SR, Gross JR, Pilch BZ, Goodman ML. Laryngeal tuberculosis as manifested in the decades 1963–1983. Laryngoscope 1987;97:848–50. [10] Hunter AM, Millar JW, Wightman AJ, Horne NW. The changing pattern of laryngeal tuberculosis. J Laryngol Otol 1981;95:393–8. [11] Ormerod FC. Tuberculosis of the upper respiratory tract. London: Staples Press Ltd.; 1939. [12] Galietti F, Giorgis GE, Gandolfi G, Astesiano A, Miravalle C, Ardizzi A, et al. Examination of 41 cases of laryngeal tuberculosis observed between 1975– 1985. Eur Respir J 1989;2:731–2. [13] Shin JE, Nam SY, Yoo SJ, Kim SY. Changing trends in clinical manifestations of laryngeal tuberculosis. Laryngoscope 2000;110:1950–3. [14] Bailey CM, Windle-Taylor PC. Tuberculous laryngitis: a series of 46 patients. Laryngoscope 1981;91:93–100. [15] Lim JY, Kim KM, Choi EC, Kim YH, Kim HS, Choi HS. Current clinical propensity of laryngeal tuberculosis: review of 60 cases. Eur Arch Otorhinolaryngol 2006;263:838–42. [16] Hiraide F. Tuberculosis in the otolaryngological regions. Jibiinkoka 1977;49:973–84 [in Japanese]. [17] Iwahashi T, Mochizuki R, Yamamoto K, Yamashita M, Muta H. Three cases of laryngeal tuberculosis: type tendencies and medical considerations. Nippon Kikan Shokudoka Gakkai Kaiho 2013;64:14–20 [in Japanese]. [18] Sasaki Y, Yamagishi F, Suzuki K, Yasuda J, Mori N, Satoh N, et al. Twelve cases of laryngeal tuberculosis. Kekkaku 1991;66:733–8 [in Japanese].
310
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
[19] Wang CC, Lin CC, Wang CP, Liu SA, Jiang RS. Laryngeal tuberculosis: a review of 26 cases. Otolaryngol Head Neck Surg 2007;137:582–8. [20] Ling L, Zhou SH, Wang SQ. Changing trends in the clinical features of laryngeal tuberculosis: a report of 19 cases. Int J Infect Dis 2010;14:230–5. [21] Bhat VK, Latha P, Upadhya D, Hegde J. Clinicopathological review of tubercular laryngitis in 32 cases of pulmonary Kochs. Am J Otolaryngol 2009;30: 327–30.
[22] Kolappan C, Gopi PG. Tobacco smoking and pulmonary tuberculosis. Thorax 2002;57:964–6. [23] Myerson MC. Smoker’s larynx. Ann Otol Rhinol Laryngol 1950;59:541–6. [24] Dye JA, Adler KB. Effects of cigarette smoke on epithelial cells of the respiratory tract. Thorax 1994;49:825–34. [25] Iseman MD. A clinician’s guide to tuberculosis. Philadelphia: Lippincott Williams & Wilkins; 2000.
Contents lists available at ScienceDirect
Auris Nasus Larynx journal homepage: www.elsevier.com/locate/anl
Laryngeal tuberculosis: A report of 17 cases Masashi Kurokawa a,*, Ken-ichi Nibu b, Kei-ichi Ichimura c, Hiroshi Nishino d a
Department of Otolaryngology, Osaka Prefectural Medical Center for Respiratory and Allergic Diseases, Japan Department of Otolaryngology-Head and Neck Surgery, Kobe University Graduate School of Medicine, Japan c Department of Otolaryngology, Ishibashi General Hospital, Japan d Department of Otolaryngology-Head and Neck Surgery, Jichi Medical University, Japan b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 December 2014 Accepted 7 February 2015 Available online 3 March 2015
Objective: The aim of this study was to reduce misdiagnosis and inappropriate treatment of laryngeal tuberculosis. Methods: Between April 2009 and March 2013, 1660 inpatients with tuberculosis were treated at the Osaka Prefectural Medical Center for Respiratory and Allergic Diseases. Seventeen of these patients were diagnosed with laryngeal tuberculosis. Criteria used for diagnosis were findings of histologic examination of biopsy specimens (2 patients) or clinical response of granuloma to antituberculosis therapy (15 patients). Medical records were used for this retrospective study. Results: Patients’ age ranged from 30 to 84 years with an average of 51.4 14.0 years (mean SE); nine were men and eight were women. The most frequent chief complaint was hoarseness (n = 15). Thirteen patients had persistent cough before they showed otolaryngologic symptoms. Endoscopic findings were categorized into five types of lesions: perichondritic (n = 6), ulcerative (n = 6), granulomatous (n = 6), polypoid (n = 1) and nonspecific inflammatory (n = 1). Laryngeal tuberculosis showed as a single lesion in one patient and as multiple lesions in the other patients. The most commonly involved site was the true vocal cord (n = 16, 94.1%), while 16 patients showed radiographic evidence of active pulmonary tuberculosis. These results indicate that deterioration of the lesions tended to result in the occurrence of multiple lesions. Conclusion: While morbidity in tuberculosis has been decreasing in Japan, a significant number of patients still has laryngeal tuberculosis in association with advanced pulmonary tuberculosis. If a patient has a protracted cough, laryngeal tuberculosis should be taken into consideration for differential diagnosis. Pulmonary imaging study should be performed for early diagnosis and to prevent the infection from spreading. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tuberculosis Larynx Perichondritic lesions Ulcerative lesions Granulomatous lesions Polypoid lesions
1. Introduction In the pre-antibiotic era, laryngeal tuberculosis used to be common and usually was associated with advanced cavitary pulmonary lesions. The incidence of laryngeal tuberculosis with pulmonary tuberculosis was about 40% [1,2]. Many patients with laryngeal tuberculosis died at that time due to severe odynophagia and dyspnea caused by laryngeal edema and granulation. After the development of chemotherapy, improvements in living standards
* Corresponding author at: Department of Otolaryngology, Kishiwada City Hospital, 1001, Gakuhara-Cho, Kisiwada-City, Osaka 596-8501, Japan. Tel.: +81 072 4445 1000; fax: +81 072 441 8812. E-mail address: [email protected] (M. Kurokawa). http://dx.doi.org/10.1016/j.anl.2015.02.012 0385-8146/ß 2015 Elsevier Ireland Ltd. All rights reserved.
and the spread of public health, laryngeal tuberculosis has become quite rare and now occurs in less than 1% of tuberculosis cases [3]. However, tuberculosis still remains a major global health problem. In 2012, an estimated 8.6 million people developed tuberculosis and 1.3 million died from the disease, including 320,000 deaths among the HIV-positive population [4]. In Japan, even though prevalence of tuberculosis has been decreasing every year since 1999, there are still more than 21,000 new patients in each year with a prevalence of tuberculosis of 16.7 per 100,000 populations in 2012 [5]. This figure is still higher than that of most developed countries. Thus, Japan belongs to tuberculosis middle burden country. In addition, AIDS and other immunosuppressive diseases or treatments also have increased and can thus be expected to further increase the incidence of tuberculosis [6,7]. However, laryngeal tuberculosis is often misdiagnosed and it thus takes time to determine the appropriate treatment,
306
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
resulting in the spread of infection, which can become a public health problem [8]. To reduce misdiagnosis and inappropriate treatment of laryngeal tuberculosis, we examined the clinical characteristics of laryngeal tuberculosis. 2. Methods Between April 2009 and March 2013, 1,660 inpatients with tuberculosis were treated at Osaka Prefectural Medical Center for Respiratory and Allergic Diseases. Seventeen of these patients were diagnosed with laryngeal tuberculosis. Criteria used for diagnosis were findings of histologic examination of biopsy specimens (2 patients) or clinical response of granuloma to antituberculosis therapy (15 patients). Patients’ medical records were retrospectively reviewed to obtain data regarding their age, presenting symptoms, site and type of the lesions, and findings of chest X-ray and CT scan, sputum smear, sputum culture, biopsy, polymerase chain reaction (PCR) tests, and QuantiFERON test (QFT). The distinguishing features of laryngeal tuberculosis will be discussed and compared with those reported elsewhere in Section 4. The data were also analyzed for primary diagnosis, patient’s delay (interval between onset of the disease and first visit to a doctor) and doctor’s delay (interval between first visit to a doctor and diagnosis as tuberculosis). The data were analyzed with Dr. SPSS II (SPSS Inc., Chicago, IL, USA), and presented as mean standard deviation (SD). An independent sample t-test was used to determine p-values. Count data were analyzed with Fisher’s exact test and p < 0.05 was considered statistically significant. 3. Results 3.1. Demography and past history The patients’ age ranged from 30 to 84 years with an average of 51.4 14.0 years (mean SE). Nine patients were men with an age range of 30–84 years (average 52.9 15.0 years) and eight were women with an age range of 33–68 years (average 50.0 12.1 years). There were two patients (11.8%) with diabetes (type 1: one, type 2: one), two with mental disorders (schizophrenia: one, social withdrawal: one), and one patient (5.9%) each with bronchial asthma and laryngeal cancer. Nine patients (52.9%) had a history of smoking. 3.2. Endoscopic findings
Fig. 1. Laryngoscopic findings. (A, B) Perichondritic lesions. (A) At start of therapy. (B) 5.5 months later. Erosion in the whole larynx and swelling of left arytenoid and the epiglottis was observed in 66-year-old man. (C, D) Ulcerative lesions. (C) At start of therapy. (D) At improvement. Whitish ulcerative lesion of both true vocal was observed in 44-year-old man. (E, F) Granulomatous lesions. (E) At start of therapy. (F) 4 months later. Granuloma-like swelling in both true vocal cords was observed in 51-year-old man.
Table 1 Findings of endoscopic laryngeal examination by category. Endoscopic findings
The various laryngoscopic findings were categorized into five different types of lesions: perichondritic (n = 3, 17.6%) (Fig. 1A), ulcerative (n = 6, 35.2%) (Fig. 1C), granulomatous (n = 6, 35.2%) (Fig. 1E), polypoid (n = 1, 5.9%), and nonspecific inflammatory (n = 1, 5.9%) (Table 1). Laryngeal tuberculosis manifested as a single lesion in only one patient, and the other 16 patients (94.1%) showed multiple lesions, with nine having severe lesions (perichondritic and ulcerative). 3.3. Symptoms and involved laryngeal subsites The most frequent chief complaint was hoarseness (n = 15, 88.2%), followed by odynophagia (n = 7, 41.2%) and abnormal sensation syndrome of the throat (n = 4, 23.5%). Thirteen patients (76.5%) had persistent cough before manifesting otolaryngologic symptoms. The infected subsites were true vocal cord (n = 16, 94.1%), arytenoids (n = 6, 35.3%), false vocal cord (n = 4, 23.5%) and epiglottis (n = 4, 23.5%). Deterioration of the lesions tended to be resulting in the involvement of multiple lesions (Table 2).
Single
Multiple
Total
%
Perichondritic lesions Ulcerative lesions Granulomatous lesions Polypoid lesions Nonspecific inflammatory lesions
0 0 1 0 0
3 6 5 1 1
3 6 6 1 1
17.6 35.2 35.2 5.9 5.9
Total
1
16
17
Epiglottis
True vocal cord
False vocal cord
Arytenoid
3 0 1 0 0
3 5 6 1 1
3 0 1 0 0
1 4 1 0 0
100
Table 2 Tuberculous sites involved in larynx. Endoscopic findings Perichondritic lesions Ulcerative lesions Granulomatous lesions Polypoid lesions Nonspecific inflammatory lesions
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
3.4. Imaging diagnosis and sputum examination Sixteen patients showed radiographic evidence of active pulmonary tuberculosis (94.1%) and chest radiographic findings were normal for only one patient even on HRCT (high-resolution CT). The sputum smear test was positive for tuberculosis for 16 patients (94.1%). Ten patients had highly contagious infection (58.8%). Results of the sputum culture and polymerase chain reaction (PCR) test were available for 17 patients; all of whom showed positive results (Table 3). 3.5. Pathologic examination and blood test (QFT) Two patients underwent a biopsy; both of whom were pathologically diagnosed with laryngeal tuberculosis. The QuantiFERON1 (QFT) test was used for two younger patients (30 and 33 years old); both of whom were positive for QFT.
307
Table 3 Results of various investigative procedures for 17 cases of laryngeal tuberculosis (PCR: polymerase chain reaction test, QFT: QuantiFERON test).
Positive Negative No record Sensitivity (%)
Chest X-ray
Chest CT
Smear test
Culture
PCR
Histology
QFT
16 1 0 94.1
16 1 0 94.1
16 1 0 94.1
17 0 0 100
13 0 4 100
2 0 15 100
2 0 15 100
Table 4 Diagnoses by otorhinolaryngologists and general practitioners. Otorhinolaryngologist Laryngeal tuberculosis Pulmonary tuberculosis Chorditis Polypoid vocal cord Glottic edema Laryngeal papilloma Common cold
(N = 8)
General practitioner
(N = 9)
2 1 1 1 1 1 1
Pulmonary tuberculosis Bronchitis Laryngeal tumor Mycoplasma pneumonia
4 2 2 1
3.6. Primary diagnosis and patient’s delay, doctor’s delay, and total delay Before coming to our Center, eight patients (47.1%) visited an otorhinolaryngologist (ENT group) and nine patients (52.9%) visited a general practitioner (GP group). The first diagnoses made by these doctors are summarized in Table 4. Patient’s delay (interval between onset of the disease and first visit to a doctor) was 181.9 113.5 days for the ENT group, and 100.6 73.9 days for the GP group. Doctor’s delay (interval between first visit to a doctor and diagnosis of tuberculosis) was 35.4 31.6 days for the ENT group, and 20.3 30.4 days for the GP group. The corresponding values for total delay (interval between onset of the disease and diagnosis of tuberculosis) were 217.3 110.7 and 120.9 60.7 days. Total delay for the ENT group was significantly longer than for the GP group (p < 0.05). Mean patient’s delay was 171.7 137.3 days for the patients with perichondritic lesions, 127.3 85.5 days for those with ulcerative lesions, 160.2 83.9 days for those with granulomatous lesions, 60 days for those with polypoid lesions, and 60 days for those with nonspecific inflammatory lesions. These values for patient’s delay were not significantly different. In relation to endoscopic findings, doctor’s delay was 3 2.8 days for perichondritic type, 20.8 22.4 days for ulcerative type, 41.3 37.5 days for granulomatous type, 54 days for polypoid type and 30 days for nonspecific
inflammatory type. Doctor’s delay for perichondritic lesions was significantly shorter than for granulomatous lesions (p < 0.05). 3.7. Smoking Nine patients (52.9%) had a history of smoking. Patient’s delay was 90.6 72.6 days for smokers and 193.1 93.4 days for nonsmokers, while doctor’s delay was 16.9 27.3 days for the former and 39.3 28.6 days for the latter. Both types of delay were longer for smokers than for nonsmokers. Patient’s delay for smokers was significantly shorter than for nonsmokers (p < 0.05) (Fig. 2). 3.8. Treatment and outcome For patients less than 80 years old, the standard treatment regimen lasted 6 months (2HREZ/4HR:INH/RFP/EB/PZA 2 + INH/ RFP 4) and for patients 80 years old and older, it lasted 9 months (2HRE/7HR:INH/RFP/EB 2 + INH/RFP 7). However, treatment had to be extended for 11 patients due to side effects for four cases, intractable disease for four cases, diabetes mellitus for one case, multiple drug-resistant tuberculosis for one case, and side effects and diabetes mellitus for one case. Ultimately, all patients attained complete remission.
Fig. 2. Difference of days for delay between smokers and nonsmokers.
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
308
Laryngeal tuberculosis also showed a favorable response to the treatment in all cases (Fig. 1A–F). Improvement of symptoms occurred within several weeks from the initiation of medical treatment. Complete resolution of laryngeal lesions was obtained within several months except for two patients with perichondritic type lesions. Of the five patients with simultaneous tracheal tuberculosis, three showed progressive tracheal stenosis in spite of medication. One of them underwent tracheostomy, one underwent tracheoplasty and one remained with complete pulmonary atelectasis. 4. Discussion 4.1. Demography and past history Thaller and Hunter reported tuberculosis is most common in the fourth to sixth decades [9,10], and this accords with the age
range for our study population. The predominance of males has long been a feature of most reports, although the male-to-female ratio varies from 1.5:1 to 9:1 [11,12]. Diabetes has been considered as a significant risk factor as also found in our series. The statistics for tuberculosis in Japan show that in 2012, 14.3% of all tuberculosis patients had diabetes and that figure has been increasing [5]. Two patients in our series had severe mental disorders, and their patient’s delay was 1 year and 9 months, while their doctor’s delay was only 1 day, suggesting that mental disease is a risk factor for the spread of tuberculosis infection. Although there were no patients with HIV among our study population, 13% of patients with tuberculosis were complicated with HIV infection in the world. In particular, approximately 43% of patients with tuberculosis in Africa were complicated with HIV infection in 2013 [4]. In fact, tuberculosis has become a major worldwide issue for patients with HIV.
Fig. 3. Decision tree for laryngeal tuberculosis. Continuous arrow: positive findings; dashed arrow: negative findings.
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
309
4.2. Endoscopic findings
4.6. Treatment outcome
The most common infection subsite was identified as the true vocal cord 40 years ago and this was also true for our series [13– 17]. As a result, the most common chief complaint has been hoarseness for patients of all ages [13–15]. As for type of appearance, about half of our patients had severe laryngeal lesions (perichondritic or ulcerative lesions) and multiple subsites were involved in most of the patients. In the pre-antibiotic era, granulomatous type tuberculosis accounted for only 10% of all laryngeal tuberculosis in the 1940s. During the period 1960–1980, however, granulomatous lesions reportedly accounted for more than 60% [16–18], probably owing to the development of antibiotics and laryngoscopy. However, more than half of the lesions in our series showed severe forms such as perichondritic and ulcerative lesions, and this ratio surpassed that reported by Sasaki et al. [18]. These results suggest that there is a tendency for patients with tuberculosis to consult medical practitioners when they are at a more advanced stage.
Laryngeal tuberculosis is known to show favorable antituberculous response [9,19]. Indeed, in most cases the symptom improve within several weeks and the larynx nearly recovers its original condition within several months. However, surgical treatment, although rarely used, may be necessary if the disease results in airway obstruction or laryngeal fibrosis [19]. We do not biopsy larynx lesions once lung tuberculosis has been confirmed, since a biopsy is not appropriate or necessary in the presence of a large number of bacteria. When a benefit agent is administered for 2 weeks, the infectivity of tuberculosis can be reduced by a factor of 10,000 [25]. Thus, when the response to tuberculosis treatment remains poor during the first several weeks, a laryngeal biopsy for differential diagnosis of laryngeal cancer should be considered.
4.3. Image diagnosis and sputum examination There are two major theories to account for infection associated with laryngeal tuberculosis [14,19]. The first theory is known as the bronchogenic theory and posits that the larynx is infected by the direct spread of large numbers of bacilli from lung tuberculosis. The other theory, known as the hematogenous and lymphogenous theory, states that tubercle bacilli spread via the lymphogenous and hematogenous routes from an early infectious lesion. A study by Auerbach found that 698 (92.3%) of 700 autopsied cases of laryngeal tuberculosis had advanced pulmonary tuberculosis [1]. While it has been reported that primary laryngeal tuberculosis has been increasing in recent years [8,13,20], the majority of cases show positive findings on X-rays in not only our study but also in others [14,19,21]. Those studies support that bronchogenic route is the main cause of infection. 4.4. Initial diagnosis and delay Total delay for the patients in the ENT group was significantly longer than that for those in the GP group. It is a fact that otorhinolaryngologists tend to approach the disease from the viewpoint of laryngeal findings, while general practitioners tend to do so on the basis of imaging-based evidence. The fact that the initial diagnosis made by otorhinolaryngologists varied widely, strongly suggests the importance of chest X-ray for correct diagnosis. Indeed, 16 of the 17 patients in our study had lung tuberculosis that could be detected by regular chest X-ray. 4.5. Smoking In our study, patient’s delay and doctor’s delay for smokers were shorter than those for nonsmokers. Nevertheless, endoscopic findings for smokers showed more progressive stages than those for nonsmokers. While several reports have dealt with the correlation between smoking and lung pulmonary tuberculosis [22], as far as we know no report has been reported on the correlation between smoking and laryngeal tuberculosis. Smoking affects the larynx mucous membrane through chronic stimulation and leads to diminished defensive capability of the larynx [23,24]. These considerations suggest that the laryngeal mucous membrane inflamed by smoking may be more likely to be affected by the transairway infection from tuberculosis and thus to result in more progressive laryngeal tuberculosis.
5. Conclusions While morbidity of tuberculosis has been decreasing in Japan, there are still numerous patients with laryngeal tuberculosis associated with advanced pulmonary tuberculosis. If the patient has a protracted cough in addition to hoarseness, not only macroscopic examination of the larynx, but also an early imaging study too is required as summarized in Fig. 3. Quite a few patients with laryngeal tuberculosis associated with advanced pulmonary tuberculosis seem to consult an otorhinolaryngolist for the initial diagnosis. However, provision of an early pulmonary imaging study and treatment to prevent the infection from spreading are also of vital importance. Conflict of interest None of the authors have any conflict of interest, financial or otherwise. References [1] Auerbach O. Laryngeal tuberculosis. Arch Otolaryngol 1946;44:191–201. [2] Harbersohn SJ. The treatment of laryngeal tuberculosis. J Laryngol 1905;20: 630–7. [3] Willams RG, Douglas-Jones T. Mycobacterium marches back. J Laryngol Otol 1995;109:5–13. [4] World Health Organization (WHO). Global tuberculosis report; 2013. [5] Tuberculosis year book. Tokyo, Japan: Japan Anti-Tuberculosis Association; 2013. [6] Burns JL. Laryngeal tuberculosis. J Otolaryngol 1993;22:398. [7] Yencha MW, Linfesty R, Blackmon A. Laryngeal tuberculosis. Am J Otolaryngol 2000;21:122–6. [8] Kandiloros DC, Nikolopoulos TP, Ferekidis EA, Tsangaroulakis A, Yiotakis JE, Davilis D, et al. Laryngeal tuberculosis at the end of the 20th century. J Laryngol Otol 1997;111:619–21. [9] Thaller SR, Gross JR, Pilch BZ, Goodman ML. Laryngeal tuberculosis as manifested in the decades 1963–1983. Laryngoscope 1987;97:848–50. [10] Hunter AM, Millar JW, Wightman AJ, Horne NW. The changing pattern of laryngeal tuberculosis. J Laryngol Otol 1981;95:393–8. [11] Ormerod FC. Tuberculosis of the upper respiratory tract. London: Staples Press Ltd.; 1939. [12] Galietti F, Giorgis GE, Gandolfi G, Astesiano A, Miravalle C, Ardizzi A, et al. Examination of 41 cases of laryngeal tuberculosis observed between 1975– 1985. Eur Respir J 1989;2:731–2. [13] Shin JE, Nam SY, Yoo SJ, Kim SY. Changing trends in clinical manifestations of laryngeal tuberculosis. Laryngoscope 2000;110:1950–3. [14] Bailey CM, Windle-Taylor PC. Tuberculous laryngitis: a series of 46 patients. Laryngoscope 1981;91:93–100. [15] Lim JY, Kim KM, Choi EC, Kim YH, Kim HS, Choi HS. Current clinical propensity of laryngeal tuberculosis: review of 60 cases. Eur Arch Otorhinolaryngol 2006;263:838–42. [16] Hiraide F. Tuberculosis in the otolaryngological regions. Jibiinkoka 1977;49:973–84 [in Japanese]. [17] Iwahashi T, Mochizuki R, Yamamoto K, Yamashita M, Muta H. Three cases of laryngeal tuberculosis: type tendencies and medical considerations. Nippon Kikan Shokudoka Gakkai Kaiho 2013;64:14–20 [in Japanese]. [18] Sasaki Y, Yamagishi F, Suzuki K, Yasuda J, Mori N, Satoh N, et al. Twelve cases of laryngeal tuberculosis. Kekkaku 1991;66:733–8 [in Japanese].
310
M. Kurokawa et al. / Auris Nasus Larynx 42 (2015) 305–310
[19] Wang CC, Lin CC, Wang CP, Liu SA, Jiang RS. Laryngeal tuberculosis: a review of 26 cases. Otolaryngol Head Neck Surg 2007;137:582–8. [20] Ling L, Zhou SH, Wang SQ. Changing trends in the clinical features of laryngeal tuberculosis: a report of 19 cases. Int J Infect Dis 2010;14:230–5. [21] Bhat VK, Latha P, Upadhya D, Hegde J. Clinicopathological review of tubercular laryngitis in 32 cases of pulmonary Kochs. Am J Otolaryngol 2009;30: 327–30.
[22] Kolappan C, Gopi PG. Tobacco smoking and pulmonary tuberculosis. Thorax 2002;57:964–6. [23] Myerson MC. Smoker’s larynx. Ann Otol Rhinol Laryngol 1950;59:541–6. [24] Dye JA, Adler KB. Effects of cigarette smoke on epithelial cells of the respiratory tract. Thorax 1994;49:825–34. [25] Iseman MD. A clinician’s guide to tuberculosis. Philadelphia: Lippincott Williams & Wilkins; 2000.
