The prevalence of diabetes in a series of patients with subglottic stenosis

Ear, Nose & Throat Journal, Nov, 2007 by Sandra L. Ettema, Todd A. Loerhl, Robert J. Toohill, Albert L. Merati

Abstract

Diabetes, which is present in 4 to 8% of adults in the United States, is a risk factor for surgical failure in laryngotracheal airway operations. We conducted a retrospective study to characterize a population of patients with subglottic stenosis--including the prevalence of diabetes, which has not been widely reported. We performed a retrospective chart review of 30 patients--22 women and 8 men, aged 17 to 77 years (mean: 47.5)--with subglottic stenosis who had presented to our facility between July 2001 and June 2004. Diabetes was present in 5 patients (16.7%); the prevalence of diabetes in our study was not significantly different from regional population-adjusted norms (8%). Although higher-grade stenosis was significantly more common in the diabetic patients than in the nondiabetic patients (p < 0.05), we were unable to conclude that diabetes plays an independent role in the development of subglottic stenosis. We intend to conduct a meta-analysis to assess the role that diabetes plays as both a risk factor for and an obstacle to the treatment of airway stenosis.

Introduction

Airway stenosis continues to be a dominant clinical entity in laryngology, and its management can be fraught with complications and setbacks. Most patients with airway stenosis acquired their obstruction after being intubated during the treatment of a major medical illness. Headway has been made in the prevention of subglottic stenosis (SGS) through the popularization of smaller, and thus safer, endotracheal tubes and a movement toward early tracheotomy. Nevertheless, the clinical scenario in which the otolaryngologist must manage a medically compromised patient with an impaired airway remains all too common.

While most cases of refractory airway stenosis can be managed with endoscopic techniques, many patients undergo open resection. Several authors have noted that diabetes mellitus is an independent risk factor for failure in open airway surgery. (1-3) This is true for resections with end-to-end anastomoses as well as for augmentation surgery, such as airway expansion with rib grafting. Yet despite the significance of these observations, the actual prevalence of diabetes in airway stenosis patients has not been widely reported. Therefore, we conducted a study to examine a cohort of patients with airway stenosis (in this case SGS) and determine the prevalence of diabetes. We chose SGS because it was the most common type of stenosis in our database of airway patients and because luminal obstruction in SGS is easily classified. (4)

Patients and methods

Our study population was drawn from a database of airway stenosis patients that had been compiled in the Division of Laryngology at the Medical College of Wisconsin. A search of the database for ICD-9 code 478.74 (stenosis of larynx) turned up 108 patients. Patients were then grouped by their clinically dominant stenosis; the most common of these was SGS (n = 32). Of these 32 patients, 2 were excluded from the study: 1 was excluded because his obstruction had occurred secondary to a pleomorphic adenoma, and the other because of subglottic amyloidosis. Our final group of 30 patients was made up of 22 women and 8 men, aged 17 to 77 years (mean: 47.5). As a group, the women were slightly younger than the men (mean ages: 46.3 years and 50.9, respectively). All patients had been seen between July 2001 and June 2004.

From the medical charts, we retrospectively recorded each patient's age and sex, the presence or absence of existing diabetes and gastroesophageal reflux disease (GERD), the location and nature of any concomitant stenoses other than SGS, and the patient's tobacco history. We then compared the prevalence of diabetes in our group with existing norms in an age-matched population.

In addition, we assigned each patient an etiology for his or her stenosis whenever possible:

* Patients who had no history of intubation were classified as having idiopathic stenosis.

* Patients who had undergone only routine intubation--for example, an uncomplicated endotracheal intubation during elective surgery--were not classified as idiopathic. If no other cause was evident, their etiologic classification was none assigned.

* A classification of prolonged intubation was assigned to patients who had undergone an endotracheal intubation (with or without mechanical ventilation) for more than 24 hours.

* A classification of MVA was assigned to patients who had experienced laryngeal and/or tracheal trauma during a motor vehicle accident.

* Patients who had undergone a tracheotomy with tube placement were classified as such.

* A GERD classification meant that the patient had signs and/or symptoms of abnormal reflux.

Finally, we assigned each patient a degree of stenosis according to the Myer-Cotton grading system. (4) Although this system was developed to evaluate pediatric SGS, it is valid for adults because it is a straightforward assessment of the degree of luminal compromise: grade I, up to 50% obstruction; grade II, from 51 to 70% obstruction; grade III, more than 70% obstruction with any detectable lumen; grade IV, no lumen.