Dental injury is well-recognized as a potential complication of laryngoscopy and tracheal intubation. However, the frequency, outcomes, and risk factors for this problem have not been documented in a well-defined patient population.
The authors analyzed the dental injuries of 598,904 consecutive cases performed on patients who required anesthetic services from 1987 through 1997. Dental injuries were defined as perianesthetic events (those occurring within 7 days) that required dental interventions to repair, stabilize, or extract involved dentition or support structures. A 1:3 case-control study of 16 patient and procedural characteristics was performed for cases that occurred during the first 5 yr of the study. Conditional logistic regression was used for data analysis.
There were 132 cases (1:4,537 patients) of dental injury. One half of these injuries occurred during laryngoscopy and tracheal intubation. The upper incisors were the most commonly involved teeth, and most injuries were crown fractures and partial dislocations and dislodgements. Multivariate risk factors for dental injury in the case control study included general anesthesia with tracheal intubation (odds ratio [OR] = 89), preexisting poor dentition (OR = 50), and increased difficulty of laryngoscopy and intubation (OR = 11).
Based on these data from a large surgical population at a single training institution, approximately 1:4,500 patients who receive anesthesia services sustain a dental injury that requires repair or extraction. Patients most at risk for perianesthetic dental injury include those with preexisting poor dentition who have one or more risk factors for difficult laryngoscopy and tracheal intubation.
DENTAL injury is a common perianesthetic event. Although there have been previous reports of the frequency of this event in small patient populations or based on survey data. [1–3] no study has determined its frequency in a large patient population or evaluated any association between patient and procedural characteristics and risk of dental injury. Therefore, to determine the frequency, outcomes, and risk factors for perianesthetic dental injuries, we identified all cases of major dental injury in a large surgical and diagnostic population and conducted a case control analysis of patient characteristics associated with this problem.
After receiving institutional review board approval, we analyzed dental injuries in 598,904 consecutive cases performed on patients of all ages who required anesthetic services at the Mayo Clinic in Rochester, Minnesota, from January 1, 1987 to December 31, 1997. Perianesthetic dental injuries were defined as injuries identified by patients within 7 days of their anesthetics that required dental consultation and led to a recommendation to repair, stabilize, or extract involved dentition or supporting structures. Cases evaluated for injury that did not require any intervention were not included. For example, patients with minor abrasions to their teeth, children in the mixed dentition ages when damage to a deciduous tooth would not affect the growth and development of their dental arches, and those patients who were concurrently undergoing dental care in which the dental injury in the perianesthetic period did not compromise their overall dental rehabilitation were not included. Patients with minor soft tissue injuries and those with temporomandibular pain dysfunction and associated myofacial pain dysfunction syndromes also were not included.
We administered a perianesthetic survey of patients more than 24 h and 7 days after they received anesthesia. This was part of an existing institutional postoperative surveillance project that has been described before.  A question to trigger the identification of any perianesthetic dental injury, "Have you noticed any change to your teeth or gums after your procedure?" was added specifically for this study. All positive responses led to further contact with the patient by a member of the investigative team. Unless it was clear that the change to teeth or gums was minor, patients were encouraged to seek dental consultation at the Mayo Clinic or with a local dentist. Interactions with these patients were continued until the outcome of the consultation was known or until the patient decided to forgo consultation. Patients who did not undergo dental evaluations and those who underwent evaluations but did not appear to the consulting dentist to have injuries meeting our definition for perianesthetic dental injury were not counted as cases. A member of our investigative team contacted all dentists who were involved in the evaluation of these patients.
All charges for dental injury consultation and repair in patients identified to have sustained a perianesthetic dental injury were obtained from patients or, with their permission, their dental providers. Because patients were not required to have their dental consultations at the Mayo Clinic, we could not obtain cost data for the dental consultations or repairs. Similarly, we did not consider costs such as time lost from work related to the dental injuries or costs to society. Instead, we report only the charges related to the dental injury consultation and, when performed, repair of the injury. The charges for these services were adjusted by inflation factors from the US Department of the Treasury for the years of this study (Consumer Price Index-All Urban Areas, Series Catalog CUUR0000AA0, Bureau of Labor Statistics, 1998).
We performed a 1:3 matched case control study using only those cases that occurred during the first 5 yr of the study period. Each patient was matched with three controls who underwent surgery or diagnostic procedures with anesthesia services on that same day. Based on dental injury rates from previous studies, [1–3] we projected that a 5-yr period would provide sufficient numbers of patients (n = 50) to provide a power of 80% to detect an odds ratio (OR) of 2.75 or more for risk factors predictive of dental injury (using a two-tailed [Greek small letter alpha] 0.05 level test and assuming 20% exposure of controls).
(Table 1) shows the potential risk factors included in the case control study. These 16 characteristics were chosen because they were identified or considered in speculation from previous studies to be associated with dental injuries. [1–3,5,6] Obesity was defined as noted in Table 1. Previous difficult tracheal intubation was noted if patients had a history of such an event recorded in their medical records before their anesthetics. Poor dentition was noted in patients who had preexisting dental abnormalities or disease that was identified in the preanesthetic medical records or by postanesthetic consultation with their dental providers. Rheumatoid arthritis was identified by a diagnosis made or confirmed by an internist or family practitioner at the Mayo Clinic. Diabetes was noted if patients were using insulin. Craniofacial abnormalities were present if the medical record documented that patients had genetic syndromes commonly associated with difficult tracheal intubation (e.g., Treacher-Collins or Pierre-Robin syndromes) or if the medical records noted the presence of micrognathia, mandibular hypoplasia, or similar craniofacial anomalies. Limited cervical range of motion was recorded if this problem was identified in the preanesthetic medical record.
Three intubation characteristics were collected. First, the laryngoscopy score of 0–2 (easy to intubate to very difficult to intubate) was recorded. This score has been described previously.  Second, the type of anesthesia provider who made the initial intubation attempt was identified. Types of anesthesia providers included board-certified or eligible anesthesiologists, certified registered nurse anesthetists, physicians training in anesthesiology, medical students participating in the anesthesiology program, and student nurse anesthetists. Third, the type of anesthesia provider (just described) who intubated the patient successfully was recorded.
The overall frequency of perianesthetic dental injury was calculated using data for the entire 11-yr study period. Chi-square analysis was used to compare the overall frequency of dental injury for patients receiving general anesthesia and for those who did not and for patients undergoing tracheal intubation and for those who did not. The potential risk factors for perianesthetic dental injury included in the case control study were analyzed using conditional logistic regression,  making use of the 1:3 matched-set feature of 68 patients and 204 controls. In all cases, a two-tailed P value <or= to 0.05 was considered significant. When appropriate, 95% confidence intervals were calculated to supplement the findings. All calculated means are reported with standard deviations.
During the 11-yr study period, 486,791 patients underwent 598,904 surgeries or diagnostic procedures and received anesthetic services. More than half (54.6%) of these patients were female. The mean age of this population was 48.6 +/- 16.2 yr (range, 0–110 yr). There were 367,017 general anesthetics and 231,887 procedures in which patients received either regional anesthetics or monitored anesthetic care. Twenty-four-hour and 7-day surveillance was possible in 99.8% and 98.4% of all patients, respectively.
We identified 132 cases of dental injury. These occurred in 126 patients undergoing general anesthetics and in 6 patients who received regional anesthetics or monitored anesthesia care (1:2,913 vs. 1:38,648, respectively, P < 0.001). For patients receiving general anesthetics, the injury rate in those with tracheal intubation was 1:2,805 compared with 1:7,390 for patients without tracheal intubation (P < 0.001).
The upper incisors were the most commonly involved teeth. In 13% of the cases, the injury involved more than one tooth. Most injuries were crown fractures and partial dislocations. One half of all injuries occurred during laryngoscopy and tracheal intubation, and another 23% were recognized as occurring during the case but after tracheal intubation. Only 8% of the cases were noted to occur during tracheal extubation. An exact time of injury was not determined for the remaining 19%. The mean repair cost was $782 +/- 412 (range, 88-$8,200; 1997-adjusted dollars).
From the analysis of the case control study, major risk factors for dental injury included general anesthesia (P < 0.001, OR = 89), tracheal intubation (P < 0.001, OR = 24), and preexisting poor dentition (P < 0.001, OR = 50)(Table 1). From a multivariate analysis that adjusted for the use of tracheal intubation, increased difficulty with tracheal intubation was found to be predictive of dental injury (moderate/difficult vs. easy; P = 0.002, OR = 11). Other preoperative conditions found to be predictive of dental injury included previous difficult tracheal intubation, limited cervical range of motion, and preexisting craniofacial abnormalities (Table 1). The type of anesthesia provider who tried or successfully performed the tracheal intubation was not predictive of dental injury.
This study confirms the presence of specific risk factors associated with dental injury. These risk factors, although they are not surprising, validate the appropriateness of the preanesthetic evaluation of patient characteristics such as a history of previous difficult tracheal intubation, limited neck motion, and previous head and neck surgery. The strong associations of general anesthesia with tracheal intubation and preexisting poor dentition with a greatly increased frequency of dental injury suggest that anesthesiologists should discuss this risk with appropriate patients.
Our findings provide current estimates of the frequency of perianesthetic dental injuries that necessitate dental intervention to repair, stabilize, or extract involved dentition. Previous studies have reported a variable frequency of this event, ranging from 1:8 to 1:1,900 patients. These studies are old,  involve small populations,  or are based on retrospective survey data of directors of academic training programs.  Therefore, it is not surprising that our frequency of dental injury (1: 4,537) is much lower than that reported in these previous studies.
Our study method has several potential flaws. First, the study was retrospective, which increased the risk that we missed cases. We believe that our intensive postanesthetic surveillance project and our high rate of successful surveillance at 24 h (99.8%) and 7 days (98.4%) minimized that risk. Second, patient and procedure characteristics for our case control study were identified by medical record review. In general, the characteristics that we studied were well-documented in our unique, single-unit medical record. At our institution, nearly all patients were examined preoperatively by board-certified or eligible anesthesiologists, internists, pediatricians, or family practice physicians. This detailed information likely increased the opportunity to collect data consistently on these characteristics. In addition, a single trained nurse-abstractor reviewed all medical records in the case control study to increase the consistency and reliability of our data collection.
In conclusion, the frequency of perianesthetic dental injury that required repair, stabilization, or extraction of the involved dentition was 1:4,537 patients in a large surgical and diagnostic medical practice. Patients who have the highest risk for dental injury are those undergoing general anesthesia with tracheal intubation who have preexisting poor dentition and characteristics (e.g., limited neck motion, previous head and neck surgery, craniofacial abnormalities, and a history of previous difficult tracheal intubation) likely to be associated with difficult laryngoscopy and tracheal intubation.