How applicable is the TNM 8 staging for human papillomavirus (HPV) related oropharyngeal squamous cell carcinoma (OPSCC) to a UK population of 106 patients? : A cohort comparison of the TNM 7 and TNM8 staging systems for HPV positive oropharyngeal cancer in a UK population.

PURPOSE: TNM8 introduced a new staging system for HPV positive oropharyngeal squamous cell carcinoma (OPSCC). This study aimed to investigate whether the changes made in TNM8 offer the perceived benefit in prognostication when compared to TNM7 in a specific patient population in the North East of England. METHODS: A retrospective cohort comparison study of all patients with HPV positive OPSCC (n = 106) through the Newcastle Head and Neck MDT between January 2012 to December 2014. Overall survival (OS) and Disease specific survival (DSS) data at 3 years was gathered for both TNM7 and TNM8. Log rank test was used to compare survival curves. Harrell's C-index adjusted for age and smoking status was used to assess prognostic ability of the two staging methods. RESULTS: TNM8 downstages disease (TNM7 stage IV patients n = 74, TNM8 stage IV patients n = 2) but gives a more even distribution of patients across disease stages. Survival for TNM8 stage II and III is similar. In our small cohort, the log-rank test detected differences in OS between stages for both scoring methods (TNM7 p = 0.006, TNM8 p < 0.001) and similarly for DSS (TNM7 p = 0.001, TNM8 p < 0.001). Harrell's C-index was similar for both models for OS (TNM7 0.71, TNM8 0.71) and DSS (TNM7 0.74, TNM8 0.70). CONCLUSION: TNM8 downstages disease and prognosticates well for stage I disease but does not differentiate between stage II and III disease when compared to TNM7. Further adaptation is required to address this to make TNM8 a more accurate prognostic tool.

A comparison of attitudes to laryngeal cancer treatment outcomes: A time trade-off study.

DESIGN: Time trade-off choice experiment. SETTING: Two large head and neck cancer centres. PARTICIPANTS: Patients who have received treatment for head and neck cancer and members of the head and neck cancer multidisciplinary team. MAIN OUTCOME MEASURES: Participants were asked to rank the outcome scenarios, assign utility values using time trade-off and rate the importance of survival on treatment choice. RESULTS: A total of 49 patients with head and neck cancer and 73 staff members were recruited. Chemoradiotherapy (CRT) optimal outcome was the most preferred health state (34 of 49, 69% patients, and 50 of 73, 68% staff), and CRT with complications was least preferred (27 of 49, 55% patients, and 51 of 73, 70% staff). Using time trade-off, mean utility values were calculated for CRT optimal outcome (0.73 for patients, 0.77 for staff), total laryngectomy (TL) optimal outcome (0.67 for patients, 0.69 for staff), TL outcome with complications (0.46 for patients, 0.51 for staff) and CRT with complications (0.36 for patients, 0.49 for staff). The average survival advantage required for a participant to change their preferred choice was 2.6 years. CONCLUSIONS: We have demonstrated that a significant proportion of patients with head and neck cancer and staff members would not choose CRT to manage locally advanced laryngeal cancer. Staff members rated the health states associated with laryngeal cancer treatment higher than patients who have experienced them, and this is particularly evident when considering the poorer outcomes. The head and neck cancer community should develop methods of practice and decision-making which incorporate elicitation and reporting of patient values as a central principle.

Transoral laser microsurgery versus radiation therapy in the management of T1 and T2 laryngeal glottic carcinoma: which modality is cost-effective within the UK?

OBJECTIVES: To identify the most cost-effective treatment strategy in patients with early stage (T1 and T2) cancers of the laryngeal glottis. DESIGN: A Markov decision model populated using data from updated systematic reviews and meta-analyses, with attributable costs from NHS sources. Data on local control and mortality were obtained from updates of existing systematic reviews conducted for the NICE guideline on cancer of the upper aerodigestive tract. Procedure costs were sourced from NHS reference costs 2013/14 by applying tariffs associated with the appropriate health resource group code SETTING: The UK National Health Service. POPULATION: Patients with early stage (T1 and T2) cancers of the laryngeal glottis. INTERVENTIONS: Transoral laser microsurgery (TLM) and radiation therapy (RT). MAIN OUTCOME MEASURES: Total costs, incremental costs and quality adjusted life years (QALYs) over a 10-year time horizon. RESULTS: Radiation therapy as the initial treatment strategy was found to be more expensive (£2654 versus £623) and less effective (QALY reduction of 0.141 and 0.04 in T1a and T1b-T2 laryngeal cancers, respectively) than TLM. The dominance of TLM for T1a cancers was unchanged in most scenarios modelled in sensitivity analysis. For T1b-T2 laryngeal cancers, the result changed in numerous scenarios. In probabilistic sensitivity analysis, TLM was found to have a 71% and 58% probability of being cost-effective in T1a and T1b-T2 laryngeal cancers, respectively. CONCLUSIONS: Transoral laser microsurgery is a cost-effective strategy to adopt in the management of T1a laryngeal cancers. Uncertainty remains over the optimal strategy to adopt in T1b-T2 laryngeal cancers.

Swallowing outcomes for patients with oropharyngeal squamous cell carcinoma treated with primary (chemo)radiation therapy receiving either prophylactic gastrostomy or reactive nasogastric tube: A prospective cohort study.

OBJECTIVES: There is currently no consensus on the optimal feeding route for an oropharyngeal squamous cell cancer (OPSCC) patient without pre-existing dysphagia undergoing chemoradiotherapy. This study aimed to compare the swallowing outcomes for OPSCC patients fed via either prophylactic RIG (pRIG) or reactive nasogastric tube (rNGT). DESIGN: A prospective cohort study. SETTING: Four head and neck cancer centres in the North of England Cancer Network. PARTICIPANTS: Fifty-three participants with OPSCC, on a normal diet pre-(chemo) radiotherapy. MAIN OUTCOME MEASURE: Swallowing outcomes for patients with rNGT and pRIG were compared using the following outcome measures administered prospectively at pre-treatment, three and 12 months post-treatment: MD Anderson Dysphagia Inventory (MDADI), Performance Status Scales (Normalcy of Diet), timed water swallow test. RESULTS: Twenty-three patients received pRIG while 30 were planned for a rNGT. No differences in demographics were found between the groups. The rNGT group had a clinically significant higher score on MDADI at 12 months post-treatment. No significant difference was found on a timed water swallow test or diet texture scale. CONCLUSIONS: There is no statistical difference for swallowing outcomes in either group. However, patients' in the rNGT group reported a clinically meaningful difference at 1 year, with a trend for them to do better across all measures. Neither group returned to their baseline swallowing function. Further research with a larger sample size is indicated.

Baseline swallowing measures predict recovery at 6 weeks after transoral robotic surgery for head and neck cancer.

OBJECTIVES: To explore whether pre-treatment swallowing measures predict swallowing recovery at 6 weeks after transoral robotic surgery (TORS). DESIGN: Prospective cohort study. SETTING: Tertiary care cancer centre in the United Kingdom. PARTICIPANTS: Fifty-one consecutive patients undergoing TORS for head and neck cancer, between April 2013 and February 2015. MAIN OUTCOME MEASURES: Swallowing function assessed by Performance Status Scale (PSS) Normalcy of Diet, timed water swallow test capacity scores (WST) and duration of tube feeding. RESULTS: The primary site distribution was as follows: 21 oropharynx, 8 larynx, 6 mucosectomy and 6 hypopharynx. T stages included 7 staged Tx, 21 T1-T2 tumours and 1 T3 tumour. Moderate-to-severe comorbidity was found in 45/51 patients. Mean PSS score was 83 (sd 27.54); mean WST score was 11.14 (sd 7.97). Most patients (73%) required tube feeding post-operatively, with mean tube feed duration of 18.08 days (sd 17.91); 76% resumed oral intake by 6 weeks. Pre-treatment swallow tests showed moderate negative correlation with tube feeding duration: PSS (rho 0-.430, P = .003); WST (rho 0-.503, P = 0.002). CONCLUSIONS: The majority of TORS patients resume oral intake by 6 weeks. This study shows that impaired swallowing prior to surgery correlates with post-operative duration of tube feeding and strengthens the evidence for the utility of these measures in this clinical setting.

Bile acids: a potential role in the pathogenesis of pharyngeal malignancy.

OBJECTIVE: Gastro-oesophageal reflux disease is thought to be a risk factor for head and neck malignancies. Bile acids are one of the principle components of gastric refluxate and have previously been implicated in the development of oesophageal and bowel malignancies. There is clear evidence that bile acids reflux into the laryngopharynx. Despite this, the carcinogenic properties of bile acids in this area are yet to be fully identified. We therefore investigated the potential role of bile acids in pharyngeal malignancy, through the highly conserved process of epithelial-mesenchymal transition (EMT). EMT occurs in invasion and metastasis and is a central process in the development of epithelial carcinoma. DESIGN: Translational research study. METHODS: Human hypopharyngeal squamous carcinoma FaDu cells were challenged with primary (cholic or chenodeoxycholic) and secondary (deoxycholic or lithocholic) bile acids. EMT-relevant proteins TGF-ß1 and MMP-9 were measured in the cell culture supernates at 48 h via ELISA. Cell viability was confirmed >95% via CellTiter-Blue assay. RESULTS: Significantly greater concentrations of TGF-ß1 were measured in the culture supernates of cells treated with cholic acid, deoxycholic acid and chenodeoxycholic acid. MMP-9 levels were increased in deoxycholic acid and lithocolic acid stimulations when compared to control (P < 0.05). CONCLUSION: This is the first demonstration that bile acids induce TGF-ß1 and MMP-9 in pharyngeal cells. TGF-ß1 is considered a master switch for EMT, while MMP-9 is a part of the EMT proteome which degrades basement membranes. This implies a potential role for bile acids in pharyngeal carcinogenesis through the mechanism of EMT and suggests potential novel therapeutic targets.

Trans-oral robotic assisted tongue base mucosectomy for investigation of cancer of unknown primary in the head and neck region. The UK experience.

OBJECTIVES: The diagnosis of cancer of unknown primary (CUP) in head and neck occurs when the treating clinicians have utilised all available diagnostic tests and failed to identify the origin of the disease. There is no agreed consensus on which diagnostic investigations to use, or the order in which to use them in, although broad recommendations exist. Small tumours arising in the tongue base can be below the limits of resolution of conventional diagnostic techniques. Given the difficulty in targeting the tongue base, current practice involves blind random biopsies, which leads to a variable detection rate. Robotically assisted surgical removal of the tongue base, tongue base mucosectomy (TBM) has been shown to improve diagnostic yield. This study reports the diagnostic hit rate for tongue base primaries using this technique. DESIGN: Retrospective case review. SETTING: UK Head and Neck Centres. PARTICIPANTS: Patients presenting as an unknown primary, investigated with clinical examination, PET-CT and palatine tonsillectomy. MAIN OUTCOME MEASURES: The detection of a primary site of head and neck cancer in the otherwise unknown primary tumour. RESULTS: The primary tumour site was identified in the tongue base in 53% (n=17) of patients. In 15 patients the tumour was in the ipsliateral tongue base (88%) while in two cases (12%) the tumour was located in contra lateral tongue base. CONCLUSIONS: Trans-oral robotic assisted TBM raises the possibility of identifying over 50% of tumours that would otherwise be classified as CUP. Identifying these in the contralateral tongue base has implications for treatment planning and outcome.

Refining the head and neck cancer referral guidelines: a two-centre analysis of 4715 referrals.

OBJECTIVES: To identify the set of referral criteria that will offer optimal diagnostic efficacy in patients suspected to have head and neck cancer (HNC) in the primary care setting. DESIGN: Statistical analysis of referral criteria and outcomes. SETTING: Two tertiary care cancer centres in the United Kingdom. PARTICIPANTS: 4715 patients who were referred via the fast-track system with a suspected HNC between 2007 and 2010. MAIN OUTCOME MEASURES: Parameters of diagnostic efficacy, multivariate regression model to calculate estimated probability of HNC and area under the receiver operating characteristic curve (AUROC). RESULTS: The majority of referring symptoms had a positive predictive value higher than the 3% cut-off point stated to be significant for HNC detection in the 2015 NICE recommendations. Nevertheless, our multivariate analysis identified nine symptoms to be linked with HNC. Of these, only four are included in the latest NICE guidelines. The best fit predictive model for this data set included the following symptoms: hoarseness >3 weeks, dysphagia >3 weeks, odynophagia, unexplained neck mass, oral swelling >3 weeks, oral ulcer >3 weeks, prolonged otalgia with normal otoscopy, the presence of blood in mouth with concurrent sensation of lump in throat and the presence of otalgia with concurrent lump in throat sensation. Intermittent hoarseness and sensation of lump in throat were negatively associated with HNC. The AUROC demonstrated that our model had a higher predictive value (0.77) compared to those generated using the NICE 2005 (0.69) and 2015 (0.68) referral criteria (P < 0.0001). An online risk calculator based on this study is available at http://www.orlhealth.com/risk-calculator.html. CONCLUSIONS: This study presents a significantly refined version of referral guidelines which demonstrate greater diagnostic efficacy than the current NICE guidelines. We recommend that further iterative refinements of referral criteria be considered when referring patients with suspected HNC.

Comparing the Performance Status Scale and MD Anderson Dysphagia Inventory as swallowing outcome measures in head and neck cancer: a prospective cohort study.

OBJECTIVES: To examine the relationship between the two disease-specific measures currently in use to assess swallowing outcomes following treatment in patients with head and neck cancer: the Performance Status Scale (PSS) and MD Anderson Dysphagia Inventory (MDADI). DESIGN: A prospective cohort study. SETTING: Four head and neck cancer multidisciplinary clinics in the North of England Cancer Network. PARTICIPANTS: 114 patients with cancers of the upper aerodigestive tract. MAIN OUTCOME MEASURE: Measures of swallowing function administered prospectively across 4 timepoints RESULTS: Spearman's correlation coefficients were used to measure the relationship between the two scales. There was statistically significant correlation between the two tools at pre-treatment (rs = 0.428, P < 0.000), 3 months post-treatment (rs = 0.454, P < 0.002), 6 months post-treatment (rs = 0.551, P < 0.000) and 12 months post-treatment (rs = 0.680, P < 0.000). CONCLUSION: This is the first prospective study comparing the MDADI and PSS questionnaires at multiple time points. Our study shows that these different instruments have a good relationship in measuring swallowing function in patients with head and neck cancer in short and medium term after treatment.

Transoral robotic surgery in head neck cancer management.

Head and Neck cancer can be treated by non-surgical or surgical modalities. Current surgical techniques include open surgery and transoral resections. The latter have the distinct advantage of quicker recovery and reduced hospital stay. The further down the aerodigestive tract tumours are sited, the more difficult is the transoral access, requiring techniques that combine magnification and finer instrumentation. Thus, while oral cavity tumours can be removed transorally without special equipment, the need to address oropharyngeal and laryngeal cancers led to the evolution of transoral laser microsurgery. Transoral robotic surgery (TORS) improves the visualisation, the instrumentation and the ergonomics in transoral resections and is also used primarily in the treatment of oropharyngeal and laryngeal cancer. Current evidence suggests that the oncologic and functional outcome of TORS surgery is good as speech and swallowing mechanisms are better preserved. This review will provide the reader an insight into the role of TORS in head and neck practice.

Transnasal oesophagoscopy: diagnostic and management outcomes in a prospective cohort of 257 consecutive cases and practice implications.

OBJECTIVES: To determine the success rate, patient tolerability and impact of introducing transnasal oesophagoscopy on clinical practices. DESIGN: Prospective cohort with review of electronic patient records for outcomes. SETTING: UK tertiary centre Otolaryngology Department. PARTICIPANTS: The cohort comprised of two hundred and fifty-seven patients, 128 females (50%) and 129 males (50%) with an age range of 20-91 years; mean age 59 years (sd 13.6). MAIN OUTCOME MEASURES: Success rates, indications, findings and outcomes of patients undergoing transnasal oesophagoscopy and impact on rigid examinations of the pharynx and oesophagus were also considered. RESULTS: Transnasal oesophagoscopy has a high success rate of 97%; it is well tolerated by patients, and poor views are uncommon. Pathology was detected in 44% of patients. The most common indications for transnasal oesophagoscopy were unexplained throat symptoms (50%) and dysphagia (25%). Common positive findings were hiatus hernia (7%), Barrett's oesophagus (5%), dysmotility (5%) and oesophageal candidiasis (5%). Following transnasal oesophagoscopy, 59% of patients were discharged to their referring clinician, 17% continued to undergo otolaryngology follow-up, and 13% were referred to our gastrointestinal colleagues. Following the introduction of transnasal oesophagoscopy, there was a reduction in the number of rigid examinations of the pharynx and oesophagus in the subsequent years, despite an increase in total referrals. CONCLUSIONS: Transnasal oesophagoscopy is a well-tolerated procedure that allows otolaryngologists to make management decisions on common referrals swiftly in the clinic setting avoiding unnecessary investigations, follow-up and referral.

Hospital stay following complex major head and neck resection: what factors play a role?

OBJECTIVES: Identify factors which have an impact on the length of postoperative hospital stay in patients undergoing major surgical resection for head and neck cancer. DESIGN: Retrospective study using hospital case notes. SETTING: Tertiary centre. PARTICIPANTS: One hundred and eighty-four patients who underwent 191 major head and neck resections between 2006 and 2010 were identified from the head and neck oncology database. An additional cohort of 52 patients from a second centre was used to externally validate the model. METHODS: Data collected on 13 variables including age, preoperative haemoglobin level, body mass index, timing of surgery (primary or recurrent tumour), category of surgical procedure (clean, clean-contaminated, contaminated, dirty), tracheostomy (yes or no) and duration of the operation were analysed using a multiple linear regression. RESULTS: In the 13 parameter model, the four variables that were found to significantly (P < 0.05) prolong length of stay (LOS) were surgical wound type (clean-contaminated, P = 2.19 × 10(-6) versus clean), tracheostomy (P = 0.0034), operation time (P = 0.011) and American Society of Anaesthesiologists (ASA) grades 3 and 4 (P = 0.0067) versus 1 and 2. The statistically significant variables were used to generate a preoperative submodel without the operative time and a postoperative submodel with all four variables. Overall, the best model based on adjusted R(2) was the postoperative model. When fitted on the external data set, there was no significant difference in the residuals, indicating that the models generalise across centres. CONCLUSION: Clean-contaminated wound was the most significant factor affecting the LOS, with others being ASA grades 3 and 4, longer duration of surgery and presence of a tracheostomy. Using these parameters, it is possible to predict the LOS in patients undergoing major surgical resection for head and neck cancer.

Thyroid gland invasion in total laryngectomy and total laryngopharyngectomy: a systematic review and meta-analysis of the English literature.

BACKGROUND: Advanced laryngeal and hypopharyngeal squamous cell carcinomas carry an inherent risk of invading thyroid parenchyma leading to the incorporation of a hemithyroidectomy or total thyroidectomy as part of a total laryngectomy. In some centres, thyroid gland removal occurs routinely during surgery for T3 and T4 laryngopharyngeal carcinoma. However, the incidence of invasion is low, and therefore, thyroid-sparing surgery must be considered for select cases. OBJECTIVE OF REVIEW: The primary goal of the review is to assess the true incidence of thyroid gland invasion in laryngopharyngeal carcinoma. Utilising this data we aim to identify risk factors and clinical predictors of thyroid gland invasion to facilitate in a more targeted approach in the surgical management of advanced laryngopharyngeal carcinoma. TYPE OF REVIEW: A systematic review and meta-analysis of all published data and review of case series at Newcastle upon Tyne Hospitals (NuTH). SEARCH STRATEGY: MEDLINE (1946-2012) and EMBASE (1980-2012) were searched. EVALUATION METHOD: A single reviewer conducted the systematic review with a follow-up ancestry search. Studies publishing case series of T3 and T4 laryngeal and hypopharyngeal carcinoma treated by total laryngectomy or laryngopharyngectomy and partial or total thyroidectomy, with pathological assessment for thyroid gland invasion rates were selected. Articles published prior to 1977 were excluded due to the advent of whole organ sectioning. RESULTS: The literature search identified 16 studies suitable for inclusion, with 1180 cases. The NuTH case series identified 107 patients. The overall pooled incidence of thyroid gland invasion in 1287 patients is 10.7% (95% CI 7.6-14.2). Patients with primary subglottic tumours (relative risk 7.5; 95% CI 4.3-13.0) and disease extension into the subglottis (relative risk 4.3; 95% CI 2.5-7.2) have a significantly higher relative risk of thyroid gland invasion. Radiorecurrent tumours and hypopharyngeal tumours did not have an increased risk of thyroid gland invasion. CONCLUSION: Advanced laryngeal and hypopharyngeal carcinomas involving the subglottis carry a significantly elevated risk of thyroid gland invasion compared with those that spare this subsite. The overall incidence of thyroid gland invasion is low, and therefore, thyroidectomy should be reserved for cases considered to be at risk as opposed to a being a routine measure for all total laryngectomies.

Achieving consensus in follow-up practice for routine ENT procedures: a Delphi exercise.

OBJECTIVES: To generate consensus amongst faculty members regarding follow-up practice for well-defined clinical scenarios using a Delphi exercise and to identify whether disseminating the consensus guidelines changed follow-up practice. STUDY DESIGN: Generation of consensus using a Delphi exercise and an audit of follow-up practice before and after dissemination of the resultant guidelines. SETTING: The department of Otorhinolaryngology-Head and Neck Surgery at the Freeman Hospital, Newcastle upon Tyne, UK. PARTICIPANTS AND METHODS: Panel members for this exercise included 11 consultants and two associate specialists and one co-ordinator. We identified clinical scenarios where ≥ 80% agreement existed that routine follow-up appointments should not be made and subsequently disseminated guidelines widely to all medical staff. The follow-up rates for the scenarios where consensus existed regarding follow-up practice were audited from clinic letters before and after the guidelines were disseminated. MAIN OUTCOME MEASURES: Agreement on scenarios where routine follow-up appointments should not be made was assessed using a Likert scale (1-5). RESULTS: Of 13 faculty members, 12 responded to rounds one and two, and 11 responded to round three. The Delphi exercise identified 18 clinical scenarios where there was ≥ 80% agreement on patients not routinely being followed up. Comparison of the follow-up practice prior to and after the Delphi exercise identified a reduction in follow-up for all 18 scenarios of 48%. CONCLUSION: Consensus regarding routine follow-up can be reached by using the Delphi process in ENT practice. This can translate into a real change in clinical practice. Furthermore, this process could be applied for consensus building in other related areas.

Use of gastrostomy in head and neck cancer: a systematic review to identify areas for future research.

OBJECTIVES: To perform a systematic review of studies, which investigate the role of gastrostomy tube feeding in head and neck squamous cell cancer. DESIGN: A systematic review of the English literature, identifying areas of clinical equipoise and recommendations for future studies. MAIN OUTCOME MEASURES: Varying practices in G-tube use in HNSCC, the benefits and disadvantages of G-tubes, the effect of G-tube placement on multidimensional outcomes in patients with head and neck cancer, including quality of life and health economics and to draw themes that may lend themselves to future research. RESULTS: The search identified 216 articles of which 59 were considered relevant. During treatment, a significant number of patients need enteral nutritional supplementation through nasogastric or G-tubes. Gastrostomy tubes have a good safety profile. Most clinicians agree that the use should be restricted to advanced tumours. Studies on prophylactic insertion of G-tubes do not show a consistent advantage to the nutritional status compared with NG tubes. At 1 year, G-tube retention rates vary between 10% and 30%. The presence of a feeding tube has a negative impact on quality of life. Placing G-tubes may have an adverse effect on swallowing after chemoradiation. CONCLUSIONS: There is a lack of consensus among clinicians about indications for placing G-tubes. There is sufficient equipoise in this area for further investigation of the effect of varying feeding practices on functional outcomes, quality of life and health economics.

The risk and interval to malignancy of patients with laryngeal dysplasia; a systematic review of case series and meta-analysis.

BACKGROUND: Laryngeal dysplasia is a pre-malignant condition with wide variability in rates of malignant transformation reported in the literature. The management and follow-up strategies of these lesions vary widely. OBJECTIVES: To assess the risk of and interval to malignant transformation in patients with laryngeal dysplasia, the effect of different treatment modalities on malignant transformation and the effects that risk factors such as smoking, excessive alcohol intake and histological grade may have on this. TYPE OF REVIEW: Systematic of observational studies with attempted meta-analysis. SEARCH STRATEGY: A structured search of Medline (1966 to January 2010), EMBASE (1980 to January 2010), CINAHL (1981 to January 2010) and Cochrane databases (CENTRAL, Cochrane Library, 1995 to January 2010). RESULTS: Nine hundred and forty cases from nine studies were included in the analysis. Overall malignant transformation rate was 14% (confidence interval 8, 22) and mean time to malignant transformation was 5.8 years. The malignant transformation rate is higher with increased severity of dysplasia grade - severe/CIS 30.4%versus mild/moderate 10.6% (P < 0.0002). Treatment modality did not show significant effects. CONCLUSIONS: Laryngeal dysplasia carries a significant risk of malignant transformation. This risk triples with increasing severity of dysplasia. Transformation often occurs late and is not related to the grade of dysplasia. There is little evidence, therefore, to support the early discharge of patients with mild/moderate dysplasia, which is practised by some clinicians.