The laryngeal mask airway (LMA; LMA North America, Inc., San Diego, CA) has a well-established role in the emergency and elective treatment of patients with difficult-to-manage airways (DA). In this study, the authors report their clinical experience with the intubating LMA (LMA-Fastrach; LMA North America, Inc., San Diego, CA) in 254 patients with different types of DA.
The authors reviewed the anesthetic and medical records of patients with DA in whom the LMA-Fastrach was used electively or emergently at four institutions from October 1997 through October 2000. In each case, the number of insertion and intubation attempts was recorded. Success rates for blind and fiber optically guided intubation through the LMA-Fastrach were calculated, up to a maximum of five attempts per patient.
The LMA-Fastrach was used in 257 procedures performed in 254 patients with DA, including patients with Cormack-Lehane grade 4 views; patients with immobilized cervical spines; patients with airways distorted by tumors, surgery, or radiation therapy; and patients wearing stereotactic frames. Insertion of the LMA-Fastrach was accomplished in three attempts or fewer in all patients. The overall success rates for blind and fiber optically guided intubations through the LMA-Fastrach were 96.5% and 100.0%, respectively.
The LMA-Fastrach was used successfully in a high percentage of patients who presented with a variety of DA. The clinical experience presented herein indicates that this device may be particularly useful in the emergency and elective treatment of patients in whom intubation with a rigid laryngoscope has failed and in the treatment of patients with immobilized cervical spines.
A CLOSED claims study published by the American Society of Anesthesiologists (ASA) in 1990 revealed the significance of respiratory problems in anesthetized patients. 1This study laid the groundwork for the development of the ASA Difficult Airway Algorithm, which established pathways for the care of patients with difficult-to-manage airways (DA). 2Although initially, the ASA Difficult Airway Algorithm recommended the use of the laryngeal mask airway (LMA ™; LMA North America, Inc., San Diego, CA) only in an emergency, Benumof 3subsequently identified five main pathways in the algorithm in which the LMA ™ may be used as a ventilating device and as a conduit for tracheal intubation. Using the LMA ™ electively or emergently in patients with DA now represents a recognized alternative in all the pathways of the ASA Difficult Airway Algorithm.
In 1997, the intubating LMA (LMA-Fastrach ™; LMA North America, Inc., San Diego, CA) was introduced. The LMA-Fastrach ™ retains all of the ventilatory features of the classic LMA ™ but is designed to provide a superior conduit for blind or fiberoptically guided tracheal intubations. 4,5Reports evaluating the effectiveness of the LMA-Fastrach ™ in the treatment of patients with DA have been limited. 6,7In this article, we present the cumulative experience of four institutions with the use of the LMA-Fastrach ™ in 254 patients with DA. In all cases, insertion of and intubation through the LMA- Fastrach ™ were performed by investigators with extensive clinical experience using the classic LMA ™ for the elective and emergency treatment of patients with DA. Before publication of this article, these investigators had begun using the LMA-Fastrach ™ in their clinical treatment of patients with DA because of the superior characteristics of the device as a conduit for blind or fiberoptically guided intubations.
Materials and Methods
After Institutional Review Board approval from all four institutions participating in this report, we reviewed the medical records and notes taken from the anesthetic records of patients in whom the LMA- Fastrach ™ was used emergently or electively for intubating purposes from October 1997 through October 2000. The four institutions who participated in this report were The University of Texas M. D. Anderson Cancer Center (Houston, TX), Yale University School of Medicine (New Haven, CT), Mt. Sinai School of Medicine (New York, NY), and the University of Chicago (Chicago, IL).
The investigators, each of whom had been trained in LMA-Fastrach ™ insertion and intubation techniques by the inventor of the device, Dr. Archie Brain, performed all LMA-Fastrach ™ insertions. Consequently, the same insertion and intubation techniques were used in every patient. Each of the clinicians also had previous experience with the use of the LMA-Fastrach ™ in patients with normal airways before using it in patients with DA.
The investigators used the LMA-Fastrach ™ in patients in whom intubation had previously proven to be difficult or was expected to be difficult based on the patient’s underlying disease and physical findings. A history of gastroesophageal reflux or hiatal hernia precluded the elective use of the LMA-Fastrach ™. However, these exclusion criteria did not apply when the device was used for emergency airway management in patients with unexpected DA in whom other methods of securing the airway had failed.
For each case, the number of insertion and intubation attempts and the success or failure to establish ventilation and to achieve intubation were reviewed. Each patient’s age, weight, height, and ASA physical status were obtained from the records, as were the locations where intubations were performed, the type of airway difficulty, the type of anesthesia used, and whether the Chandy maneuver was used (Dr. Chandy Verghese, Consultant in Anaesthesia and Intensive Care, Royal Berkshire Hospital, Reading, England, verbal communication, January 1998). Any complications that were documented as having occurred during LMA-Fastrach ™ insertion and intubation were also reviewed.
Insertion of the LMA-Fastrach ™
As recommended in the LMA-Fastrach ™ instruction manual, 8the patient’s head was placed in a neutral position during insertion by placing a support between the operating table and the occiput of the patient. 8After insertion of the device, ventilation was confirmed by observing chest movement and capnography. The following anesthetic techniques for insertion of the LMA-Fastrach ™ were identified.
For the majority of elective cases, in patients with no airway pathology, general anesthesia was induced with propofol (2.0–2.5 mg/kg). 9In patients whose airways had been altered by tumors, surgery, or radiation therapy, inhalation induction with sevoflurane (2–4%) was used while spontaneous respiration was maintained. For patients in whom insertion of the LMA-Fastrach ™ was performed while the patient was awake, topical anesthesia to the airway was achieved with 4% lidocaine spray to the oropharynx and 3–4 ml lidocaine, 4%, was injected through the cricothyroid membrane. Emergency insertions of the LMA-Fastrach ™ were performed in unconscious patients outside the operating room and in anesthetized patients in the operating room in whom other techniques for intubation had failed.
In patients with unstable cervical spines, LMA- Fastrach ™ insertion and intubation were performed with rigid Philadelphia collars (Philadelphia Cervical Collar Co., Westville, NJ) in place. A small portion of the foam from the chin section of the Philadelphia collar was cut out to provide adequate access to the patient’s mouth during LMA-Fastrach ™ insertion and to allow for the correct insertion technique. As determined by the neurosurgeons before using the LMA-Fastrach ™, this modification did not affect stabilization of the neck.
Intubation through the LMA-Fastrach ™
Blind intubation through the LMA-Fastrach ™ was attempted electively in patients with no intrinsic tumors distorting the airway. A nondisposable, silicone endotracheal tube (ETT; ILM Endotracheal Tube; Euromedical, Lake Zurich, IL) designed for blind intubation through the LMA-Fastrach ™ was used in all patients. Water-soluble lubricant was applied to the ETT before insertion into the LMA-Fastrach ™.
A useful maneuver to aid the correct positioning of the LMA-Fastrach ™ and to facilitate blind intubation was developed by Dr. Chandy Verghese (fig. 1). The Chandy maneuver consists of two steps , which are performed sequentially. The first step, which is important for the establishment of optimal ventilation, is performed by slightly rotating the device in the sagittal plane using the metal handle until the least resistance to bag ventilation is achieved. The second step of the Chandy maneuver is performed just before blind intubation and consists of using the metal handle to slightly lift (but not tilt) the LMA-Fastrach ™ away from the posterior pharyngeal wall. This facilitates the smooth passage of the ETT into the trachea. Although all of the authors were aware of the Chandy maneuver, only one used it consistently in his clinical practice.
When the LMA-Fastrach ™ is aligned with the glottic opening, little or no resistance is encountered when passing the ETT into the trachea. However, when resistance is encountered, the LMA-Fastrach ™ instruction manual 8recommends further manipulations to assist passage of the ETT. These manipulations are based on measuring how far the ETT has been advanced beyond the distal opening of the metal tube of the LMA- Fastrach ™ when resistance is encountered. That distance can easily be calculated by counting the 1-cm markings above a black transverse mark on the ETT, which represents the point where the ETT emerges from the distal opening of the metal tube of the LMA-Fastrach ™. Based on the manual’s recommendations, the investigators used the following guidelines for subsequent intubation attempts.
1. If resistance was felt after advancing the ETT 3 cm beyond the distal opening of the LMA-Fastrach ™ tube, the device was too small, and a larger LMA-Fastrach ™ was used.
2. If resistance was felt within 1 cm when trying to advance the ETT, the device was too large, and a smaller LMA-Fastrach ™ was used.
3. If resistance was felt 2–2.5 cm beyond the distal opening of the LMA-Fastrach ™ tube, the epiglottis had become down-folded during insertion and was not within reach of the epiglottic elevating bar (EEB). In this case, the LMA-Fastrach ™ was partially withdrawn and reinserted.
To avoid airway trauma, force was not used to overcome resistance when advancing the tracheal tube. In addition, the number of intubation attempts was limited to five, after which intubation was completed with the aid of a fiberoptic bronchoscope (FOB; Olympus LF-2, Olympus BF XT-30; Olympus America, Inc., Melville, NY).
Fiberoptically assisted intubation was chosen as the initial method of intubation in patients with airway tumors and in those whose airways had been altered by previous surgery or radiation therapy. During fiberoptically guided intubations through the LMA-Fastrach ™, the FOB was initially kept inside the ETT. However, after the tip of the ETT reached and lifted the EEB, the FOB was advanced through the tip of the ETT and passed between the vocal cords into the trachea. The ETT was then advanced over the FOB until tracheal intubation was achieved.
For each attempt to insert or intubate through the LMA-Fastrach ™, success rates were calculated for all patients and for each type of DA. To analyze factors that influenced the success rates for insertion of and intubation through the LMA-Fastrach ™, we used the Pearson chi-square test to compare independent proportions and reported the differences in proportions in terms of the odds ratio with corresponding 95% confidence intervals. A P value less than 0.05 was considered significant.
The LMA-Fastrach ™ was used for tracheal intubation on 257 occasions, which included 246 intubations (95.7%) performed in 243 patients in an operating room, 5 intubations (1.9%) in an emergency room, 4 intubations (1.6%) in a radiology suite, and 2 intubations (0.8%) in an intensive care unit (table 1).
According to the anesthesia and medical records, each patient had at least one of the following characteristics: (1) a failed intubation associated with a grade 4 laryngoscopic view as defined by Cormack and Lehane 10; (2) an immobilized cervical spine; (3) a distorted airway secondary to tumor, surgery, or radiation therapy; or (4) the presence of a stereotactic head frame that limited access to the airway.
The LMA-Fastrach ™ was used in patients during general anesthesia on 200 occasions (77.9%) and in patients during topical anesthesia and sedation in 51 cases (19.8%). On six occasions (2.3%), no anesthesia was used because the patients were unconscious.
The LMA-Fastrach ™ was successfully inserted on the first attempt in 228 cases (88.7%), on the second attempt in 27 cases (10.5%), and on the third attempt in 2 cases (0.8%) (table 2). There were no failures to insert the device, and ventilation was possible in all patients.
Blind intubation through the LMA-Fastrach ™ was attempted in 200 cases and was successful in 193 (96.5%). Blind intubation was achieved on the first attempt in 151 cases (75.5%). On the remaining blind intubation attempts, successful intubation was achieved on the second, third, fourth, and fifth attempts in 28 (14.0%), 7 (3.5%), 5 (2.5%), and 2 cases (1.0%), respectively. In seven cases (3.5%), blind intubation through the LMA-Fastrach ™ failed after five attempts, and fiberoptically guided intubation was successful on the first attempt. In 57 cases, fiberoptically guided intubation was the initial intubating technique, with a 100.0% success rate on the first attempt. There were no pharyngeal, laryngeal, or esophageal complications associated with the use of the LMA-Fastrach ™ in any of the patients included in this study.
Patients in Whom Previous Intubation Techniques Had Failed
According to the records we reviewed, in 111 cases (43.2%), previous intubations with a rigid laryngoscope had failed, and the LMA-Fastrach ™ was subsequently used emergently for intubation during general anesthesia on 23 occasions (20.7%) and electively as the primary intubating device in the remaining 88 cases (79.3%).
In the 23 cases in which the device was used emergently, intubation was performed blindly in 20 cases and with an FOB in 3 cases. Blind intubation was achieved on the first attempt in 15 cases (75%). Blind intubation was achieved on the second and third attempts in two cases (10.0%) each and on the fourth attempt in one case (5.0%). Fiberoptically guided intubations were successful on the first attempt in all three cases.
In 88 of 111 cases of patients with a history of failed intubation, the LMA-Fastrach ™ was used electively with an overall success rate for blind intubation of 92.0% (n = 81). Blind intubation was achieved on the first attempt in 56 cases (63.6%). Successful blind intubation was achieved on the second, third, fourth, and fifth attempts in 15 (17.1%), 4 (4.5%), 4 (4.5%), and 2 cases (2.3%), respectively. In seven patients (8.0%), blind intubation failed, and intubation was completed successfully with the aid of an FOB. During FOB observation in these patients, three were found to have an epiglottis that was unusually narrow and omega-shaped and could not be lifted effectively by the EEB. In the other four patients, the epiglottis appeared soft and floppy, sliding off the EEB during blind intubation attempts and causing the ETT to slide over the epiglottis into the valleculae.
Patients with Immobilized Cervical Spines
The LMA-Fastrach ™ was used in 70 cases (27.2%) of patients with unstable cervical spines that were immobilized in rigid Philadelphia collars. In 67 cases, insertion of the LMA-Fastrach ™ and intubation were performed during general anesthesia. In two cases, topical anesthesia and sedation were used. One patient was unconscious, and no anesthesia was administered. Blind intubation was achieved on the first attempt in 63 cases (92.6%); in 5 cases (7.4%), two attempts were required. In two cases, elective fiberoptically guided intubation through the LMA-Fastrach ™ was performed successfully. Of particular significance, no new neurologic deficits were observed secondary to airway management with the LMA-Fastrach ™.
The LMA-Fastrach ™ was also used in 14 patients (5.4%) whose cervical spines were immobilized by previous surgical fusion. Blind intubation was successful in 13 patients: on the first attempt in 10 patients (76.9%), on the second attempt in 2 patients (15.4%), and on the third attempt in 1 patient (7.7%). In one patient, elective fiberoptically guided intubation through the LMA- Fastrach ™ was performed successfully.
Patients with Distorted Airways
The LMA-Fastrach ™ was used on 50 occasions (19.5%) in patients with distorted airway anatomies. In two patients with intrinsic tumors involving the airway, the LMA-Fastrach ™ was used emergently after other methods of securing the airway had failed. In these patients, intubation through the LMA-Fastrach ™ was achieved with an FOB. In eight patients, tumors caused extrinsic airway compression. In two of these patients, intubation through the LMA-Fastrach ™ was performed blindly, with successful intubation achieved on the first attempt in one patient and on the second attempt in the other patient. In six patients with extrinsic airway compression, intubation was performed successfully through the LMA-Fastrach ™ with initial use of an FOB. In all 40 patients with airway changes related to previous surgery, radiation therapy, or both, fiberoptically guided intubation was used electively and was successful through the LMA-Fastrach ™.
Patients Wearing Stereotactic Frames
In 12 patients (4.7%) fitted with Leksell (Elekta Instruments Inc., Norcross, GA) or Compass (Regulus Corp., Rochester, MN) stereotactic frames, the LMA-Fastrach ™ was used electively for airway management. In the first three cases, insertion was performed with the patient awake and during topical anesthesia, and intubation was achieved with fiberoptic guidance. In the remaining cases, insertion was performed during general anesthesia, and blind intubation was achieved on the first, second, and third attempts in six cases (50.0%), two cases (16.7%), and one case (8.3%), respectively.
The Chandy Maneuver
Both steps of the Chandy maneuver were applied in 73 of the 200 cases of patients in whom blind intubation through the LMA-Fastrach ™ was attempted. For patients with all types of DA, the incidence of multiple intubations was significantly lower when the Chandy maneuver was used before intubation (4 of 73 [5%]vs. 41 of 127 [32%], odds ratio = 0.12, 95% confidence interval, 0.04–0.36 multiple intubation attempts, P = 0.0001).
For patients with DA, the LMA-Fastrach ™ combines the ventilatory benefits of the classic LMA ™ with an easier and more reliable method of tracheal intubation. The LMA-Fastrach ™ can accommodate an ETT with an ID of up to 9.0 mm, and the device can easily be disengaged from the ETT when intubation has been completed.
The shape of the LMA-Fastrach ™ is based on magnetic resonance imaging studies of the human airway, ensuring a close fit with the oropharyngeal curve and optimal alignment with the glottic opening when the head is maintained in a neutral position. Our results show that this design frequently permits successful and smooth blind passage of the ETT into the trachea. To reduce the risk of airway trauma associated with the use of blind intubation techniques, a special silicone ETT (maximum ID, 8.0 mm) has been developed by Dr. Brain and is recommended for use with the LMA-Fastrach ™. This silicone tube has a low-volume cuff to minimize the risk of damage to it during intubation through the metal shaft of the LMA-Fastrach ™. However, because silicone has different physical properties than do rubber and polyvinyl chloride, the use of this low-volume silicone cuff is not associated with the increased risk of high transmucosal pressure that is encountered with the use of these other materials. 11An isolated case of esophageal perforation associated with intubation attempts through the LMA-Fastrach ™ has been described, 12but we found no evidence of esophageal or airway trauma associated with this device in 257 intubations.
Patients in whom intubation with a rigid laryngoscope fails often have a “high” or “anterior” larynx, which is classified as a Cormack-Lehane grade 4 view. Face mask ventilation may also be difficult in these patients and may further be compromised by soft tissue swelling from multiple intubation attempts. The consequence may be complete failure to perform ventilation in the patient, resulting in serious morbidity or death. 1
In all 111 cases in our study in which rigid laryngoscopy, intubation, or both failed because of a Cormack-Lehane grade 4 view, insertion of the LMA-Fastrach ™ and ventilation were successful. Although the success rate for first-attempt blind intubation in this group of patients was only 65.2%, within five attempts, it reached 92.0%. In seven patients, blind intubation failed, but intubation was successful with an FOB. Using fiberoptic guidance during intubation through the LMA-Fastrach ™ is very helpful, especially for an infrequent user of both devices. However, in situations in which an FOB may not be readily available, blind intubation through the LMA-Fastrach ™ is a viable option and with experience can be performed successfully in a high percentage of patients. Regardless of the intubation technique used, the ability to perform successful ventilation in patients with the LMA-Fastrach ™ was probably the biggest advantage of the device in this group of patients with DA.
All airway maneuvers cause some movement of the cervical spine. 13,14In patients with cervical spine instability, the challenge is to select an intubation technique that minimizes the risk of further injury associated with airway manipulation. 15The LMA-Fastrach ™ is designed to be used while maintaining the patient’s head and neck in a neutral position, and insertion of the device does not require any external manipulation of the patient’s neck. Therefore, we hypothesized that the device should be safe to use in patients with unstable cervical spines who are wearing rigid Philadelphia collars.
In a 1999 study of human cadavers, Keller and Brimacombe 16reported that pressure against the vertebrae of the cervical spine increased during LMA-Fastrach ™ insertion, and they cautioned against using the device in patients with unstable cervical spines. However, there is no clinical evidence that the pressure changes described in the cadavers would result in significant cervical spine movement during LMA-Fastrach ™ use. Kihara et al. 17showed that in 20 patients with cervical spine pathology, LMA-Fastrach ™ insertion and intubation with manual in-line stabilization were associated with only a 0.5- to 1.0-mm posterior displacement of C2–C5 vertebrae. These results are comparable with those of Hauswald et al. , 13who found that fiberoptically guided oral and nasal intubations were associated with mean cervical spine movements of 1.78 and 1.08 mm, respectively. Indeed, based on the most recent cadaveric study by Brimacombe, 18it seems that the LMA-Fastrach ™ and the fiberoptic scope are the devices least likely to cause significant cervical spine movement when used for intubation.
There have been no reports of neurologic injury associated with the use of the LMA-Fastrach ™ in patients with cervical spine pathology. 19–21We used the LMA-Fastrach ™ in 70 patients with unstable cervical spines, and none sustained a neurologic injury associated with the use of the device, despite the fact that in most of the patients, the instability involved the C2–C4 cervical vertebrae, which were closest to the LMA-Fastrach ™ and most likely to be affected. Further experience with and controlled studies of airway management with the LMA-Fastrach ™ in patients with unstable cervical spines are necessary to fully define the role of the device in these patients.
In the current study, insertions of the LMA-Fastrach ™ were performed with the Philadelphia collar in place. However, Wakeling and Nightingale 22studied the ease of insertion and intubation through the LMA-Fastrach ™ in healthy patients whose necks were immobilized with a different cervical collar (Stifneck Select; Laerdal Medical Corp., Wappinger’s Falls, NY). They found it difficult to insert the LMA-Fastrach ™ because the collar limited how wide the patient’s mouth could be opened. They also stated that with the collar in position, the larynx seemed to have moved upward and forward, making it difficult to ventilate and impossible to intubate blindly through the LMA-Fastrach ™. After the collar was removed, blind intubation was successful in all patients in their study on the first attempt through the LMA- Fastrach ™. We observed a similar experience with the Stifneck Select collar in one patient with an unstable cervical spine who was not a part of this report. 23Further research is necessary to determine how different rigid collar designs affect airway anatomy and which collar design is most suitable for use with the LMA-Fastrach ™.
Our data indicate that the LMA-Fastrach ™ may also have a role in the ventilation and intubation of patients whose cervical spines have been immobilized by surgical fusion. In the 14 such cases presented in this article, LMA-Fastrach ™ insertion, ventilation, and intubation were performed in all 14 patients during general anesthesia and without the use of a fiberoptic scope.
Intrinsic or extrinsic tumors involving the upper airway may cause significant anatomic distortion, making tracheal intubation difficult or impossible. In two patients who had general anesthesia with intrinsic tumors of the upper airway, rigid laryngoscopy and fiberoptically guided intubation failed, but ventilation and intubation with the LMA-Fastrach ™ were successful. Although we do not recommend the primary use of supraglottic airways in patients with upper airway tumors, we found that the LMA-Fastrach ™ may be useful in the emergency airway management of these patients when other techniques have failed.
In patients with airway distortion secondary to extrinsic compression, the LMA-Fastrach ™ also proved useful in establishing ventilation. In one patient with a large goiter, there was significant tracheal deviation, and fiberoptic guidance with a bronchoscope 6.0 mm in diameter was used to achieve intubation through the LMA- Fastrach ™. Wakeling 24reported a similar case in which intubation failed because the EEB of the LMA-Fastrach ™ could not be lifted with a smaller fiberoptic scope.
In patients whose neck tissues have normal elasticity, insertion and inflation of the classic LMA ™ and the LMA-Fastrach ™ cause the soft tissues of the pharynx to expand outward. However, this expansion is limited in patients with tissue fibrosis resulting from previous neck surgery or radiation therapy, making correct positioning of the LMA-Fastrach ™ more difficult. In our experience with 40 such cases, we found that in 10 patients, limited tissue elasticity required the use of a smaller LMA- Fastrach ™ than indicated by the patient’s height and weight. We also used fiberoptic guidance during tracheal intubation. Indeed, our experience suggests that in patients with primary airway tumors or airway changes secondary to therapy, fiberoptic guidance should always be used when attempting tracheal intubation through the LMA-Fastrach ™.
Airway management of patients undergoing stereotactic procedures may be difficult because access to the patient’s airway is limited. A stereotactic frame interferes with face mask ventilation, and rigid laryngoscopy may not be feasible. We used the LMA-Fastrach ™ electively in 12 patients fitted with Leksell or Compass stereotactic frames and found that after insertion of the LMA- Fastrach ™, ventilation was achieved in all patients, and intubation was successful in all through the device.
After insertion of the LMA-Fastrach ™, the optimal alignment of the device with the glottis cannot always be guaranteed. Applying the Chandy maneuver can be helpful in achieving optimal alignment of the device as reflected by a lack of resistance to bag ventilation. The best ventilating position tends also to be the best intubating position. In our hands, the use of the Chandy maneuver was associated with significantly higher rates of successful first-attempt blind intubations through the LMA- Fastrach ™. However, the usefulness of this maneuver and its impact on the success rates for blind intubation through the LMA-Fastrach ™ need to be studied further in a prospective manner.
The limitation of this study is that our data are not comparative and, of necessity, unblinded. However, we have demonstrated the safe and effective use of the LMA-Fastrach ™ in 254 patients with DA. Insertion of the LMA-Fastrach ™ and intubation through it were successful on the first attempt in a high percentage of patients with various types of DA. This represents the largest analysis to date examining the use of the LMA-Fastrach ™ in patients with DA, and success rates were similar to those reported in previous studies. 5–7,17,19Our experience also demonstrates that the LMA-Fastrach ™ may be a particularly valuable tool for the emergency or elective airway management of patients in whom other techniques have failed and in the treatment of patients with immobilized cervical spines.
The authors thank Kenneth R. Hess, Ph.D. (Associate Professor, Department of Biostatistics, The University of Texas M. D. Anderson Cancer Center, Houston, TX), for his assistance with the statistical analysis of data from this study.