Introduction
Laryngeal cancer is one of the commonest head and neck malignancies and is second in prevalence among the head and neck cancers in the USA.1 Over 2000 cases of laryngeal cancer are diagnosed annually in the UK.Reference Jones, De, Foran, Harrington and Mortimore2 The diagnosis is based on clinical findings and confirmed by biopsy of the suspicious lesion. Traditionally, biopsy is carried out under general anaesthesia. With recent advances in technology, this procedure can now be undertaken in the out-patient setting. Office-based laryngeal procedures have come a long way over the years with the advances in technology, such as distal chip endoscopes, narrow-band imaging and better local anaesthetic techniques.Reference Wellenstein, Schutte, Takes, Honings, Marres and Burns3–Reference Shoffel-Havakuk, Halperin, Yosef and Lahav5
A wide range of laryngeal procedures can be carried out under local anaesthesia in the out-patient setting. This includes laryngeal and hypopharyngeal biopsy, vocal fold injections, and laser surgery for vocal fold polyps and granulomas.Reference Rosen, Amin, Sulica, Simpson, Merati and Courey6–Reference Caffier, Nawka, Ibrahim-Nasr, Thomas, Müller and Ko8 Laryngeal procedures carried out under local anaesthesia have advantages and disadvantages. The advantages include earlier diagnosis and access to treatment, that they can be performed on patients not suitable for general anaesthesia, and that they are suitable for patients with difficult anatomy and significant co-morbidities. There may also be a pecuniary benefit for the National Health Service because an office procedure is far less resource intensive than an elective in-patient operation. The disadvantages include intolerance of the procedure and a potential inconclusive diagnosis.
The aim of this systematic review was framed using the Population, Intervention, Control and Outcomes model to assess the efficacy, safety and cost-effectiveness (outcomes) of an office-based laryngeal biopsy under local anaesthesia (intervention), in patients with laryngeal lesions requiring histological diagnosis (population) in comparison with biopsy under general anaesthetic (control).
Materials and methods
This systematic review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (‘PRISMA’) statement standards.Reference Moher, Liberati, Tetzlaff and Altman9 The published literature in PubMed, Embase, Google Scholar and Medline were searched independently by two authors (RS and SRO). The search terms and Boolean operators used were: ‘larynx’ AND ‘biopsy’ AND ‘outpatients’ OR ‘office’.
The inclusion criteria were: studies on laryngeal biopsies in the out-patient setting, English language articles, human participants and papers that were published from 1 January 2001 to 31 October 2021. The exclusion criteria were laser therapy, laryngeal injections and other therapeutic laryngeal procedures. Article bibliographies and citations were also used to identify additional articles. The last search was run on 1 November 2021. Titles and abstracts were screened to determine whether they met the inclusion criteria. Full-text publications of all studies that were not primarily excluded were obtained and read in full. Outcome measures were sensitivity and specificity of laryngeal biopsies, complications, and cost analysis. The quality of the included studies was independently assessed by authors RS and SRO. The studies were graded on level of evidence according to the Oxford Centre for Evidence-Based Medicine Levels of Evidence Working Group.10
Results
The database search identified 302 articles from Embase (n = 122), Medline (n = 82), PubMed (n = 186) and Google Scholar (n = 30). After removing duplicates and exclusion of studies, 35 studies met the inclusion criteria. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (‘PRISMA’) flowchart is shown in Figure 1. A descriptive analysis was undertaken because of heterogeneity between studies. Table 1 summarises the effectiveness and complications of out-patient laryngeal biopsy.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (‘PRISMA’) study flow chart.
Summary of effectiveness and complications of out-patient laryngeal biopsy
PPV = positive predictive value; NPV = negative predictive value
Diagnostic yield
High rates of success in obtaining a diagnosis from laryngeal biopsy under local anaesthesia have been reported in the literature. In a retrospective study by Lippert et al.Reference Lippert, Hoffman, Dang, McCulloch, Hartig and Dailey11 on 116 patients, the diagnosis was made in 97 patients, and only 2 patients did not tolerate the procedure. In a case–control study by Zalvan et al.Reference Zalvan, Brown, Oiseth and Roark12 on 26 patients who had biopsy of the larynx carried out under both local and general anaesthesia, they found an overall concordance rate of 81 per cent, of which 83 per cent were benign and 79 per cent were non-benign. In a study by Cohen & BenyaminiReference Cohen and Benyamini13 with 117 patients, 110 patients had adequate biopsy with an office-based procedure. Similarly, in a study by Wellenstein et al.,Reference Wellenstein, de Witt, Schutte, Honings, van den Hoogen and Marres14 a definitive diagnosis was obtained for 97 per cent of cases. Reasons for failure included inability to insert the flexible endoscope, abandoning the procedure because of complications such as laryngeal bleeding and supraglottic oedema, and patient intolerance. In 1 out of 201 flexible endoscopic biopsy procedures, histological classification could not be determined by the pathologist because of the superficial nature of the biopsy specimens.
Sensitivity and specificity
Many studies have looked at the sensitivity of office-based laryngeal biopsies, with rates up to 97 per cent.Reference Cha, Yoon, Jang, Lee, Lee and Wang15–Reference Chang, Lin, Hsin, Lee, Lin and Li24 Excellent rates for specificity have been reported in multiple studies ranging from 83 per cent to 100 per cent.Reference Cha, Yoon, Jang, Lee, Lee and Wang15–Reference Chang, Lin, Hsin, Lee, Lin and Li24 A retrospective review on 581 patients by Cha et al.Reference Cha, Yoon, Jang, Lee, Lee and Wang15 showed a sensitivity and specificity of 78.2 per cent and 100 per cent, respectively. The negative predictive value of malignancy diagnosed on out-patient biopsy was 87.3 per cent, and the positive predictive value was 100 per cent. They reported a high false negative rate in glottic lesions and lesions less than 1 cm of 27 per cent and 27.5 per cent, respectively.
A case cohort study performed by Hassan et al.Reference Hassan, Usman, Yousuf, Ahmad and Hirani16 on 47 patients gave a specificity of 75.6 per cent and sensitivity of 100 per cent. A similar study by Chang et al.Reference Chang, Lin, Hsin, Lee, Lin and Li24 on 390 patients reported a sensitivity and specificity of 77.8 per cent and 95.1 per cent, respectively. In the study by Cohen & BenyaminiReference Cohen and Benyamini13 on 117 patients, the sensitivity and specificity were 70.6 per cent and 96.7 per cent, respectively. Castillo Farías et al.Reference Castillo Farías, Cobeta, Souviron, Barberá, Mora and Benito18 compared the efficacy of office-based biopsy performed under local anaesthesia and operating room biopsy performed under general anaesthesia in 88 patients. This was a prospective study performed during two periods of time, with one blinded and the other one unblinded. They had a sensitivity of 81 per cent and specificity of 100 per cent. The negative predictive value was 20 per cent.
Cohen et al.Reference Cohen, Safadi, Fliss, Gil and Horowitz19 carried out a prospective study on 102 patients with laryngeal lesions who underwent office-based biopsy. An adequate sample was obtained in 96 patients. The biopsy showed carcinoma in 34 patients, carcinoma in situ in 17 patients and was benign in 45 patients. All patients with a benign and carcinoma in situ diagnosis had a direct laryngoscopy and biopsy. The sensitivity of office-based biopsy compared with direct laryngoscopy was 69.2 per cent and specificity was 96.1 per cent. Richards et al.Reference Richards, Sugumaran, Aviv, Woo and Altman20 carried out a retrospective case series on office-based biopsy on 261 patients. The sensitivity was 60 per cent and specificity was 87 per cent. A prospective study by Uys et al.Reference Uys, Pelser, Attwood, Adam, Afrogheh and Hille21 reported a sensitivity, specificity and accuracy of 77.1 per cent, 100 per cent and 97.9 per cent, respectively. A retrospective study by Saga et al.Reference Saga, Olalde, Larruskain, Álvarez and Altuna22 showed sensitivity and specificity of endoscopic in-office biopsy to be at 73 per cent and 100 per cent, respectively, and they reported positive and negative predictive values of 100 per cent and 30 per cent, respectively.
The approach for in-office flexible endoscopic laryngeal biopsy can be transnasal or transoral. A cross-sectional cohort study by Refaat et al.Reference Refaat and Negm23 looked into the difference in sensitivity and specificity of the two approaches. Sixty patients underwent out-patient biopsy: 30 underwent the transnasal approach and 30 underwent the transoral approach. The sensitivity and specificity of the transnasal approach was 26.3 per cent and 90.9 per cent, respectively. The sensitivity and specificity of the transoral approach was 95.8 per cent and 83.3 per cent, respectively.
Few studies have looked into improving sensitivity and specificity by employing narrow-band imaging. A retrospective study by Chang et al.Reference Chang, Lin, Hsin, Lee, Lin and Li24 using the narrow-band imaging technique showed sensitivity and specificity rates of 97.2 per cent and 100 per cent, respectively. The diagnostic accuracy was 98.9 per cent.
Chung-Tsung et al.Reference Pan, Lee, Fang, Li, Liao and Chen25 performed laryngeal biopsy with narrow-band imaging on patients not suitable for general anaesthesia. Nineteen patients were included in the study: 12 patients had a positive result from the initial biopsy, and 1 had a positive result after a repeat biopsy. Six patients had a benign result and were just followed up.
Complications
The literature has shown in-office flexible endoscopic laryngeal biopsy to be a safe procedure with only a few complications reported. Chang et al.Reference Chang, Lin, Hsin, Lee, Lin and Li24 mentioned that all their patients tolerated the procedure except one patient who required temporary fluid resuscitation. Cohen et al.Reference Cohen, Bishara, Trushin and Benyamini17 reported on four patients who developed complications, giving a complication rate of 1 per cent. These complications included epistaxis (n = 2), vocal fold haematoma (n = 1) and aspiration (n = 1). In the study by Wellenstein et al.,Reference Wellenstein, de Witt, Schutte, Honings, van den Hoogen and Marres14 4 patients out of 201 developed complications, which were graded according to Clavien–Dindo classification. The noted complications included laryngospasm (Clavien–Dindo grade I), anterior epistaxis (Clavien–Dindo grade I), laryngeal bleeding (Clavien–Dindo grade II) and supraglottic oedema (Clavien–Dindo grade IIIb).
A recognised advantage of in-office flexible endoscopic laryngeal biopsy is its feasibility in patients who are not fit for general anaesthesia. Mozzanica et al.Reference Mozzanica, Ottaviani, Ginocchio and Schindler26 studied 55 patients who were unfit for general anaesthetic and went on to have out-patient biopsies. Thirty-four patients were diagnosed with malignancy with a diagnostic rate of 61.8 per cent. There were no reported complications in this high-risk group.
Cost saving
The shift from an operating room to an in-office endoscopic procedure has been shown to be more cost efficient in the literature. In a study by Schutte et al.,Reference Schutte, Takes, Slootweg, Arts, Honings and van den Hoogen27 the average cost of out-patient biopsy was €105.11 as opposed to €864.83 for operating-room biopsy. Cha et al.Reference Cha, Yoon, Jang, Lee, Lee and Wang15 showed that the cost of out-patient biopsy was $50 USD compared to $1200 USD for operating-room biopsy. Castillo Farías et al.Reference Castillo Farías, Cobeta, Souviron, Barberá, Mora and Benito18 showed an annual savings of $50 140.80. Naidu et al.Reference Naidu, Noordzij, Samim, Jalisi and Grillone28 compared the cost of out-patient biopsy and in-patient biopsy on 12 patients in a retrospective study. The cost of out-patient biopsy was $2053.91, and the cost of in-patient biopsy was $9024.47, giving a cost saving of $7000 for out-patient procedures. Fang et al.Reference Fang, Li, Liao, Chiang and Chen29 performed a retrospective study on the cost effectiveness of out-patient and in-patient biopsy and its impact on the Taiwanese health insurance programme. The cost of out-patient biopsy was 1264 Taiwanese dollars compared with in-patient biopsy which was 10 913 Taiwanese dollars. Marcus et al.Reference Marcus, Timen, Dion, Fritz, Branski and Amin30 performed a retrospective cost analysis study on patients who underwent office-based biopsy and operating-room biopsy and showed the cost for these were $7000 and $11 000, respectively. They also concluded that even if the patient has to undergo operating-room biopsy after office-based biopsy, it was still found to be cost effective. A prospective study carried out by Uys et al.Reference Uys, Pelser, Attwood, Adam, Afrogheh and Hille21 showed a saving of 5343 South African Rand with out-patient endoscopic laryngeal biopsy as opposed to rigid laryngoscopy. A retrospective study by Saga et al.Reference Saga, Olalde, Larruskain, Álvarez and Altuna22 showed an 80 per cent cost reduction with out-patient endoscopic laryngeal biopsy. The cost saving results are summarised in Table 2.
Costs of out-patient laryngeal biopsy compared with in-patient biopsy
OP = out-patient; IP = in-patient
Time for diagnosis
Diagnosis using in-office endoscopic biopsy has been shown to be more time efficient, possibly leading to a better patient outcome. Lee et al.Reference Lee, Smith, Chandarana, Matthews, Bosch and Nakoneshny31 performed a retrospective study on 114 patients, of which 44 underwent out-patient biopsy. They looked at time to diagnosis and treatment between the two groups. They found the time to diagnosis for office biopsy to be 1.3 days and 17.4 days for operating-room biopsy. The time from initial consultation to treatment was 51.7 days and 44.6 days for operating-room biopsy and out-patient biopsy, respectively. Schutte et al.Reference Schutte, Takes, Slootweg, Arts, Honings and van den Hoogen27 found the time to treatment from first patient contact in out-patient biopsy was 27 days as opposed to 41.5 days for operating-room biopsy. The time to diagnosis and treatment is summarised in Table 3.
Time to diagnosis or treatment
OP = out-patient; IP = in-patient
Discussion
Summary of results
We analysed results on the sensitivity and specificity of the biopsy results of the studies included. The population group ranged from 19 patients to 581 patients. The sensitivity ranged from 60 to 97.2 per cent and specificity ranged from 83.3 to 100 per cent. Use of narrow-band imaging has improved the sensitivity and specificity rates up to 97.2 and 100 per cent, respectively. The complication rate ranged from 0 to 2.35 per cent in the reported studies, with the largest study on 581 patients demonstrating a 0 per cent complication rate. Most studies showed that out-patient laryngeal biopsy provides a significant cost saving, with one study showing an annual cost saving of $50 140.80. Overall, a percentage cost saving ranging from 36 to 95.8 per cent has been demonstrated. Out-patient endoscopic laryngeal biopsies have also been shown to considerably reduce time to both diagnosis and treatment.
Quality of evidence
Most of the studies were observational studies with a level of evidence from 2 to 4. There were no randomised, controlled studies identified. Several studies had a small sample size with numbers less than 100, including those performed by Lippert et al.,Reference Lippert, Hoffman, Dang, McCulloch, Hartig and Dailey11 Zalvan et al.Reference Zalvan, Brown, Oiseth and Roark12 and Hassan et al.Reference Hassan, Usman, Yousuf, Ahmad and Hirani16 The retrospective review on 581 patients by Cha et al.Reference Cha, Yoon, Jang, Lee, Lee and Wang15 had the limitations of the possibility of under-diagnosis of pre-malignant lesions and that not all the benign lesions were re-biopsied.
A bi-centric prospective study performed by Chabrillac et al.Reference Chabrillac, Espinasse, Lepage, Uro-Coste, Dupret-Bories and De Bonnecaze32 compared the results of laryngeal biopsy using white light and narrow-band imaging with direct laryngoscopy and biopsy. The narrow-band imaging result contributed to an upgrade of clinical T-staging in 4.8 per cent of patients, which had an impact on the type of treatment administered to one patient who was switched from exclusive radiotherapy to an organ preservation protocol. However, all the biopsies were only performed in the operating theatre under general anaesthetic.
Bensoussan and AndersonReference Bensoussan and Anderson33 carried out a survey in Canada for those performing out-patient laryngeal procedures, focusing on the safety aspects of practice. An electronic survey questionnaire was sent to 22 practising otolaryngologists and 16 responded. The results of the survey showed that 20 to 30 per cent of otolaryngologists did not have access to medical resources in the event of any complication, such as laryngospasm or allergic reaction. Also, it is worth mentioning that only the study by Wellenstein et al.Reference Wellenstein, de Witt, Schutte, Honings, van den Hoogen and Marres14 graded complications according to the Clavien–Dindo classification.
There are currently no guidelines in laryngology for office-based laryngeal procedures in terms of patient selection, cardiac screening, local anaesthetic dosage, stopping anti-coagulants and post-procedure monitoring. The limitation of our study was that none of the studies included were of level 1 evidence. However, given the nature of the procedure, it may be difficult to do a randomised, controlled study. The faster access to diagnosis and treatment, resulting in better prognosis and survival, through office-based biopsy remains uncertain.Reference Lippert, Hoffman, Dang, McCulloch, Hartig and Dailey11 The longitudinal study of long-term outcomes may clarify this.
Practical issues
The success of the procedure depends on proper planning, good communication with the patient, adequate local anaesthesia, patient selection and skill of the surgeon performing the procedure.Reference Wellenstein, Schutte, Marres, Honings, Belafsky and Postma34 Patients with a suspicious lesion on the larynx naturally become anxious about having a procedure under local anaesthetic. If they tolerate the initial examination well, then with appropriate anaesthetic and skill, the procedure can be completed successfully. The absolute contraindication for an office-based procedure is a compromised airway.Reference Zalvan, Brown, Oiseth and Roark12 The relative contraindication was anti-coagulant use and patients who cannot tolerate the procedure because of the gag reflex. However, the study carried out by Fritz et al.Reference Fritz, Peng, Born, Cerrati, Verma and Wang35 did not show any complications for those who had office-based laryngeal procedures and were on anti-coagulants.
A variety of local anaesthetic techniques have been described. Endoscopic office-based procedures are usually well tolerated in up to 90 per cent of patients, but this drops to 70 per cent in patients with uncontrolled gag reflex and anxiety.Reference Young, Smith, Sulica, Krishna and Rosen36 Effective local anaesthetic of the larynx and the pharynx is therefore important in completing the laryngeal procedure successfully. The choice of local anaesthetic is 4 per cent lignocaine or 2 per cent Cetacaine®. Adequate decongestion of the nasal cavity is also needed, and this can be achieved with 25 per cent Neo-Synephrine® spray.Reference Woo37 However, different centres use different local anaesthetic. Cohen et al. used 2 per cent lignocaine mixed with 0.05 per cent oxymetazoline hydrochloride for the nose, 10 per cent lignocaine spray for the soft palate and 2 per cent lignocaine for the larynx that was administered through the working channel.Reference Cohen, Safadi, Fliss, Gil and Horowitz19
Superficial biopsies may result in false negative results.Reference Naidu, Noordzij, Samim, Jalisi and Grillone28 If clinical suspicion is high, it is recommended to take multiple and deeper biopsies. With the advancement of narrow-band imaging, a targeted biopsy can be taken, which will yield a higher positive result. Narrow-band imaging can identify early changes in the mucosal and submucosal vascular pattern and help in early diagnosis of superficial lesions.Reference Filauro, Paderno, Perotti, Marchi, Garofolo and Peretti38 There are five mucosal aspects on narrow-band imaging, with I to IV being non-malignant and Va to Vc being malignant according to Ni et al.Reference Ni, He, Xu, Gao, Lu and Yuan39 Zhou et al.Reference Zhou, Zhang, Guo, Nie, Zhu and Ma40 carried out a meta-analysis on the diagnostic accuracy of narrow-band imaging in laryngopharyngeal cancers and found a sensitivity of 91 per cent and specificity of 91.5 per cent. Qi et al.Reference Qi, Yu, Zhao, Jin, Sun, Liu, Cheng and Zhang41 performed an analysis of 3675 patients who attended the ENT department with different laryngeal symptoms. They used both white light and narrow-band imaging to examine and biopsy the suspected lesions. Malignant lesions show a demarcated brownish area with scattered brown spots on narrow-band imaging. Biopsy of 168 patients with features of malignancy on narrow-band imaging showed only two false negatives. This was because of overlying thick keratin and necrosis.
Schimberg et al. carried out a literature review on the cost of office-based laryngeal procedures and found the cost varied between €57 to €110 for out-patient biopsy and €822 to €1101 for operating-room biopsy.Reference Schimberg, Wellenstein, van den Broek, Honings, van den Hoogen and Marres42 The present review has also demonstrated a percentage cost saving ranging from 36 to 95.8 per cent in office-based laryngeal procedures over operating-room biopsy.
Conclusion
Office-based laryngeal procedures have been increasingly performed with the advent of newer technology. The analysis of the published data clearly shows the benefit on cost, safety and efficacy of out-patient laryngeal biopsy under local anaesthetic. Even though the sensitivity is low in some of the studies, the use of narrow-band imaging improved sensitivity and specificity for laryngeal biopsy.
Competing interests
None declared
