Hoarseness: Can Evidence-based Medicine and Practical Experience Get Along?

August 6, 2010

Professor, Director


As we move toward a world where medical treatments are judged not by pizzazz but by efficacy and cost-effectiveness, evidence-based medicine (EBM) has become the new thrust of medical education and practice.  Along with it, interest in clinical practice guidelines (CPGs) has burgeoned.  When developed according to protocol, these guidelines serve to minimize unnecessary care and encourage high-yield investigations and interventions.  There are times, however, when clinical acumen and common sense can both run afoul of such guidelines.  Many consider this to be the case with the guideline for Hoarseness recently published in Otolaryngology-Head and Neck Surgery.

The Clinical Practice Guideline: Hoarseness (Dysphonia)1 is one of the five developed in conjunction with the American Academy of Otolaryngology-Head and Neck Surgery (AAO).  Past guidelines have dealt with otitis media with effusion (2004), adult sinusitis (2007) and benign paroxysmal positional vertigo  (BPPV) (2009).  The AAO’s stated purpose is “to improve the quality of care for patients with hoarseness based on current best evidence” and at the outset, it defines hoarseness as a symptom and dysphonia as a diagnosis.  A working group of otolaryngologists, other physicians, speech pathologists, voice teachers and allied health advocates was convened to survey and analyze the available evidence for evaluation and management of the spectrum of voice disorders.  The panel then made eleven statements which were classified as strong recommendations, recommendations or options according to the literature on which they were based.  The guideline was unveiled in the September edition of the Otolaryngology-Head & Neck Surgery and at a miniseminar at the 2009 AAO Annual Meeting.  At this event, this new CPG was reviewed and each statement explained to the audience.  Controversy abounded at this meeting as many seasoned veterans expressed dissatisfaction with the process as well as the product.

The guideline was recently the subject of another panel. This time, it was at the American Broncho-Esophagological Association   (ABEA) Annual Meeting held at the 2010 Combined Otolaryngology Spring Meeting (COSM).  In this forum, strong critics of the AAO hoarseness guideline reviewed its statements and explored some of its weaknesses.  Two camps have developed within the ENT community- those who support the guideline and those who see it as an incomplete reflection of the literature which bucks both common sense and accumulated experience. The panel began with a review of the nature of physicians’ thought process and the differences between clinical trials and everyday practice and emphasized the unique way the experienced practitioner synthesizes available data and the unique circumstance of the patient to generate a plan of action.  The panel also expressed concern that the dispassionate analysis of incomplete/inadequate literature would not serve patients well and that accumulated experience should not be eliminated from the management of these often complicated disorders.

The disagreement between these camps begins with the first statement about the diagnosis of hoarseness.  While we all may use the ICD code for dysphonia, we also recognize that it is a symptom of an underlying disorder of laryngeal or vocal function rather than the end-game of our investigation.  Other CPGs have dealt with disease entities rather than symptoms.  The ABEA panel universally decried the CPG’s first statement as a misstep for this reason, asserting that this places the entire guideline on shaky ground since the literature on symptoms is not as precise as that for diagnoses.

Statement 3 re-ignites the conflict as it sets a 3-month time frame by which the larynx should be examined.  Some have likened this to allowing up to 3 months before the Orthopedic Surgeon should examine the knee of a patient who presents with joint pain and an inability to weight-bear.  When an organ is perceived to malfunction, it should be directly examined as soon as feasible in the decision-making process.  While the CPG posits that the dysphonic larynx may be examined at any time and emphasizes factors that should trigger prompt evaluation, the statement is open to easy misinterpretation by those inexperienced in the care of diseases of the head and neck region.

The CPG does also include ideas generally believed to align well with standard clinical experience.  Statement 2 emphasizes the importance of the history for determining the cause of dysphonia and points out certain historical factors such as prior cervical surgery, prolonged intubation and esophagectomy which involve risk to the recurrent laryngeal nerve.  Statement 4 recommends against the routine use of imaging for dysphonia prior to direct examination of the larynx with at least mirror laryngoscopy. Statements 5 and 6 refer to the use of anti-reflux medications and steroids, respectively.  In the case of each, the analysis shows that the risks of these medications can easily outweigh the benefits and thus, they should be used selectively for specific pathology.  Statement 7 admonishes against the routine use of antimicrobial agents until a definitive bacterial cause is identified while #8 recommends that the larynx be examined prior to recommending voice therapy, an intervention which was regarded as useful in the right circumstances.  Statements 9 and 10 advocate appropriate use of surgical therapy and chemo-denervation for specific conditions. The final statement offers the option of educating patients about the nature of their voice problem and preventative measures that may be taken to avoid future occurrences.

The CPG says more about the state of evidence in otorhinolaryngology than anything else.  The statements were based on the group’s analysis of aggregate literature. Of the 11, only 2 are based on level A evidence (well-designed randomized controlled trials or diagnostic studies performed on a population similar to the guideline’s target population) while 3 rise to level B (randomized controlled trials or diagnostic studies with minor limitations or overwhelmingly consistent evidence from observational studies).  The CPG concludes by recognizing this trend in the extant otorhinolaryngology literature and identifying areas in need of structured research on which we can base our decisions. We should remember, however, that EBM is population-based and cannot always be extrapolated to every individual.  Each patient we see is an individual with particular circumstances to which we must respond.  Since these guidelines are tempting fodder for payors to use when determining pay structures, we should work to make sure that they are fair and balanced…with our thumbs on the scale in favor of our patients’ best interests.  In the end, just as we temper justice with mercy, we must learn to practice evidence-based medicine that is tailored to the needs of our individual patients.