Low-Tube-Current Multidetector CT for Children with Suspected Extrinsic Airway Compression

Transcription

1 Preeyacha Pacharn 1,2 Stacy A. Poe 3 Lane F. Donnelly 1,3 Received February 20, 2002; accepted after revision May 15, Department of Radiology, Children s Hospital Medical Center, 3333 Burnet Ave., Cincinnati, OH Address correspondence to L. F. Donnelly. 2 Present address: Department of Radiology, Siriraj Hospital and Mahidol University, Bangkok, Thailand Department of Pediatrics, Children s Hospital Medical Center, Cincinnati, OH AJR 2002;179: X/02/ American Roentgen Ray Society Low-Tube-Current Multidetector CT for Children with Suspected Extrinsic Airway Compression OBJECTIVE. The purpose of our study was to review the technical success achieved using low-tube-current multidetector CT for the evaluation of children with suspected extrinsic airway compression and to evaluate the need for sedation during this procedure. MATERIALS AND METHODS. We reviewed all CT examinations performed for the evaluation of extrinsic airway compression during the first year after installation of a multidetector CT scanner at a pediatric hospital. We recorded the technical parameters including tube current, kilovoltage, slice thickness, mode of study, sedation technique, and amount of contrast material and noted which postprocessing techniques were applied. Studies were evaluated for timing of contrast bolus, image quality, motion artifact, need for sedation, and the diagnoses made. RESULTS. Fifty-four studies were performed in 50 patients (30 boys, 20 girls; age range, 15 days to 17 years; mean age, 2.4 years). The mean tube current was 52.2 ma (range, ma). Thirty-four studies (63%) were performed without sedation: 12 with sedation administered under supervision of the radiologist, six with general anesthesia supervised by an anesthesiologist, and two in patients who arrived in the radiology department already intubated. Imaging quality was excellent in 35 studies (65%), diagnostic in 19 studies (35%), and poor in none. Motion artifact was present on several slices in two examinations (4%). Contrast medium administration was well-timed in 49 studies (91%), early in three studies (5%), and late in two studies (4%). Airway abnormalities were detected in 26 (48%) of the studies and included extrinsic compression by vascular anomalies (n = 14) or nonvascular masses (n = 5) and intrinsic airway disease without extrinsic compression (n = 7). CONCLUSION. Evaluation for extrinsic compression of the airway in children can be accomplished using a low-tube-current multidetector CT protocol; in most pediatric patients, the examination can be performed without sedation. E xtrinsic airway compression in pediatric patients has various causes. In addition to the classic vascular rings, airway compression may result from a variety of enlarged or malpositioned vascular structures as well as from nonvascular masses adjacent to the airway [1]. MR imaging has proven to be an optimal imaging modality for evaluating pediatric patients with such anomalies because of its excellent intrinsic contrast and multiplanar capabilities [1 3]. However, the need for sedation accompanying pediatric MR imaging and the associated increased risks in patients with a compromised airway has led to interest in alternative imaging modalities. Multidetector CT allows the rapid acquisition of volumetric data sets. This rapid acquisition time reduces the need for sedation in pediatric patients [4]. By using a low-tube-current technique [5] and rapid table speed, radiation exposure (the major disadvantage of CT in pediatric patients) can also be minimized. The purpose of our study was to examine the technical success and need for sedation in the use of low-tube-current multidetector CT for the evaluation of extrinsic tracheobronchial compression in children. Materials and Methods We reviewed all CT examinations performed for evaluation of suspected extrinsic airway compression in pediatric patients during the first year after installation of a multidetector CT scanner at our institution. All patients were referred from pediatric subspecialists (pediatric otolaryngology or pediatric pulmonology) for suspected extrinsic airway compression and showed stridor with or without related symptoms (noisy breathing, wheezing, retraction, or AJR:179, December

2 Pacharn et al. recurrent infection). Most patients had undergone endoscopic evaluation that showed or suggested extrinsic compression, often pulsatile in nature. Permission to review this material was obtained from our institutional review board. All patients were imaged on a multidetector CT scanner (LightSpeed QXi, General Electric, Milwaukee, WI). Imaging was performed according to a standard protocol based on the following technical parameters: 120 kvp; weight-based low tube current [5]; 0.8-sec gantry rotation; slice thickness of mm, depending on patient size; high-quality mode (pitch 3:1); and 2 ml/kg of IV iodinated contrast medium with a scan delay of 20 sec after completion of contrast medium administration. The tube current used for an infant-sized patient (most of the patients in our study) was 30 ma. IV contrast medium was administered either by rapid hand injection or by power injection at 2 3 ml/sec. Multiplanar or threedimensional reconstructions (created by reformatting images in the coronal or sagittal plane or generating volume-rendered images of the airways and vascular structures) were performed at the discretion of the supervising pediatric radiologist. The sedation technique used for each examination was recorded. Patients either received no sedation or underwent conscious sedation supervised by the radiologist or general anesthesia supervised by an anesthesiologist; two patients arrived in the radiology department already intubated because of clinical factors not related to preparation for imaging. We used analysis of variance to evaluate whether differences existed among the mean ages of the sedated, nonsedated, and general anesthesia groups. Sedation was administered according to our department s structured sedation program [6], which meets external guidelines including those of the American Academy of Pediatrics [7]. The patient s history was reviewed and a focused physical examination was performed, after which informed consent was obtained from the parents or legal guardians. The choice of sedation drug was based on patient age; oral chloral hydrate at mg/kg was typically used in patients younger than 18 months. IV pentobarbital, in an initial dose of 3 mg/kg with repeated dosage up to 7 mg/kg, was used in those older than 18 months. During conscious sedation and recovery after sedation, patients were observed directly, and physiologic indicators including level of consciousness, blood oxygen saturation as measured by pulse oximetry, patient color, respiratory rate, and ECG tracing were monitored and recorded. If the child was considered to be at excessive risk for conscious sedation, general anesthesia was performed by the anesthesiology department. The decision regarding which sedation technique to use was made before the procedure by the performing radiologist and referring physicians. Two pediatric radiologists reviewed and evaluated all CT studies, and conclusions were made by consensus. The reviewers were not aware of endoscopic or surgical findings at the time the studies were reviewed. The studies were evaluated for timing of contrast bolus (early, well-timed, or late), image quality (excellent, diagnostic, or poor), motion artifact (present or absent), and the usefulness of reformatted multiplanar or three-dimensional images. A Fisher s exact test was used to determine whether the grading of image quality differed among the sedated, nonsedated, and general anesthesia groups. Reformatted images were considered useful if they revealed findings not seen on the axial images or increased the reviewer s confidence in a finding suspected on axial images. The value of reformatted images in communicating findings to the referring physician was not considered in this study. Imaging findings related to airway abnormalities were recorded and categorized as intrinsic airway abnormalities or extrinsic airway compression. Extrinsic compression was considered to be present if airway narrowing was seen in conjunction with an abnormal vascular or nonvascular structure immediately abutting the area of airway narrowing. An intrinsic airway abnormality, such as tracheomalacia, stenosis, or complete tracheal rings, was considered to be present if airway narrowing was seen in the absence of an abnormal structure immediately abutting the area of airway narrowing. If the caliber of the trachea was small and round in configuration, complete tracheal rings were considered to be present. To evaluate the diagnostic accuracy of the CT findings, we compared CT results with findings at endoscopy or surgery for all patients in whom such procedures were performed. Results Fifty-four studies were performed in 50 pediatric patients during the first year after a multidetector CT scanner was installed at our institution. Our study included 30 boys and 20 girls who ranged in age from 15 days to 17 years (mean age, 2.4 years). Thirty-three (66%) of the patients were 2 years old or younger. The mean tube current used was 52.2 ma (range, ma). Thirty-four studies (63%) were performed without sedation. Conscious sedation was used in 12 studies (22%). Six studies (11%) were performed with the patient under general anesthesia. Two studies (4%) were performed in children who arrived in the CT suite already intubated because of clinical factors not related to preparation for imaging. The mean age of the 34 patients imaged without sedation was 2.8 years, with an age range from 15 days to 17 years. Twenty-seven of the 34 nonsedated patients were younger than 5 years old. The mean age of the 12 patients sedated by radiology staff was 1.6 years, with an age range from 1 month to 6 years 6 months. The mean age of the eight patients who underwent general anesthesia (including the two who arrived intubated) was 2.1 years, with an age range of 1 month to 12 years. We found no statistically significant difference in age between the sedation groups (p = ). No patient who underwent CT without sedation had to undergo repeated CT with sedation because of poor image quality. The timing of the contrast bolus was optimal in 49 (91%), early in three (5%), and late in two (4%) of studies. IV contrast material was administered to all patients. Image quality was classified as excellent in 35 (65%) of studies and diagnostic in 19 (35%) of studies. The primary reason that the 19 studies were considered diagnostic rather than excellent was the presence of increased noise, which was related to the low tube current and rapid table speed. Motion artifact was present on several images in two (4%) of the examinations. No studies were considered technically nondiagnostic, and no major difference in image quality existed among the sedation groups. Of the 34 studies performed without sedation, 21 (62%) of studies were graded as excellent and 13, as diagnostic; one patient had motion artifact on several images. Of the eight CT studies performed with the patient under anesthesia, five (63%) were graded as excellent and three as diagnostic; one patient had motion artifact on several images. Of the 12 studies in which patients were sedated by radiology staff, nine (75%) of examinations were graded as excellent and three as diagnostic; none of these patients had motion artifact. No statistically significant differences were found between the sedation groups in the quality of the examinations (p = ). Multiplanar or three-dimensional reformatted images were created in 16 (30%) of studies and were considered diagnostically helpful in four (25%) of these cases (Fig. 1). Twenty-six (48%) of studies showed abnormalities of the airway, and the airway was normal in 28 studies. Seven patients had intrinsic airway abnormalities, including tracheomalacia or stenosis in six and complete tracheal rings in one (Fig. 2). Extrinsic tracheal compression was detected in 19 patients: 14 had airway compression associated with vascular causes including double aortic arch in three (Fig. 3), right arch with aberrant left subclavian artery in three, innominate artery compression in four, pulmonary sling in one, and other vascular causes in three. Five patients had airway compression caused by nonvascular masses (Fig. 4). There were no cases in which findings on CT were shown to be inaccurate at endoscopy or surgery. Of the 26 studies performed on patients with airway abnormalities, all patients were evaluated endoscopically either before or after CT was performed. All had endoscopic findings consistent with the findings shown on CT. The 19 CT studies that showed extrinsic compression revealed fixed narrow AJR:179, December 2002

3 CT of Extrinsic Airway Compression in Children Fig. 1. Additional information provided by reformatted CT images in 5-month-old girl with herniated liver causing cardiac displacement and resultant left main bronchus compression between heart and descending aorta. A, Axial CT scan acquired at lung window level shows normal caliber of right main bronchus (arrowhead) and severe narrowing of left main bronchus (arrow). B, Axial CT scan acquired at mediastinal window level shows anterior hernia of liver (L) with posterior displacement of cardiac structures. C, Sagittal reconstruction CT image shows herniated liver (L), posterior displaced heart, and compression of left main bronchus (arrow) between displaced heart and descending aorta. Symptoms of airway compression resolved after surgical hernia repair. A ing at the corresponding anatomic level. Of the seven studies with CT findings of intrinsic tracheal disease, six showed tracheomalacia (nonfixed narrowing) and one revealed tracheal rings at endoscopy; all of these findings were suggested by findings on CT. Twenty of the CT studies that depicted airway abnormalities were in patients who had undergone surgical procedures. Six of these were tracheotomies only (six with tracheomalacia and one with innominate artery compression syndrome). Surgical procedures confirmed the CT findings in the remaining 13 cases. Of the 28 CT examinations that did not show airway abnormalities, none were of patients who went on to surgery. Nineteen CT examinations with negative findings were in patients who underwent endoscopy. Of these, nine patients underwent endoscopy before CT; these patients had questionable endoscopic findings of airway compression that led to their referral for CT. Because the CT examinations in these nine patients showed normal findings, they were treated conservatively. AJR:179, December 2002 B C Pediatric airway compression caused by vascular anomalies can be responsible for respiratory distress and can require surgical intervention [2]. It is important to show the cause and degree of tracheal narrowing in such patients so that appropriate management and possible surgical intervention can be planned. The development of cross-sectional imaging has provided the capability to obtain detailed anatomic information and has facilitated accurate diagnosis, precise surgical planning, earlier treatment, and better outcomes [8]. MR imaging has traditionally been the modality of choice to evaluate pediatric patients with airway obstruction. The multiplanar capability and high-quality intrinsic contrast afforded by MR imaging allow excellent visualization of Discussion Fig year-old boy with complete tracheal rings. CT shows small caliber and round appearance of trachea (arrows). 1525

4 Pacharn et al. the tracheobronchial tree, aortic branches, and compressing vessels without exposing patients to ionizing radiation. MR imaging has been shown to accurately depict the level, severity, and cause of airway compression [1 3]. However, MR imaging is a costly modality, and because of the relatively long time required for image acquisition, young children usually require sedation. Extrinsic airway compression typically presents early in life; therefore, the need for sedation arises in most of the pediatric patients who undergo MR imaging for this reason. The risks associated with sedation are generally considered to be higher for patients with a compromised airway [9 12]. The paradoxical problem is that those patients who undergo imaging studies to evaluate for airway compression and who require sedation are at increased risk for depression of airway reflexes as a result of that sedation. The development of multidetector CT scanners that allow the acquisition of extended volume data from each gantry rotation has reduced scanning time to 20 33% of that required for single-section helical CT [4]. The A Fig month-old boy with double aortic arch. A and B, CT scans acquired at mediastinal (A) and lung (B) window levels show double aortic arch with left (L) and right (R) arches with extrinsic tracheal compression (arrow, A and B). Fig year-old boy with plexiform neurofibroma and previously undiagnosed neurofibromatosis type 1 who presented with stridor from airway compression. CT scan shows infiltrative superior mediastinal mass (M) compressing trachea (arrows). speed of multidetector CT has facilitated optimal contrast enhancement and decreased the need for sedation. The rate of sedation use in multidetector CT has reportedly been reduced to one third of that required with single-detector helical CT [13]. Reducing the need for sedation is of particular importance in patients with compromised airways. Sixty-three percent of our studies were performed without sedation. Only 22% of our studies required conscious sedation. Fifteen percent were performed while the patient was under general anesthesia; this included 11% in patients for whom other sedation methods were deemed inadequate and 4% in patients who arrived in the department already intubated for other reasons. The mean ages in the sedated and nonsedated groups were similar. Radiation exposure is the major disadvantage of CT, and doses in children that are equivalent to those used in adult CT protocols have been suggested to be associated with a small increase in the risk of future malignancy [14]. The relationship between radiation exposure and increased cancer incidence is disproportionately high in infants as compared with adults [14]. Because a large percentage of children imaged for airway compression are infants, the issue of radiation exposure from CT is particularly important. Radiation exposure can be minimized by using a weight-based low tube current and rapid table speed [5]. Radiation dose is directly proportional to tube current [5, 15]. The 30 ma tube current used in chest studies of infants at our institution involves a radiation dose of only one fifth that used in a typical (180 ma) adult protocol [15]. Also, this series used a pitch of 3:1, resulting in one third the radiation dose that would be encountered on the same equipment at a pitch of 1:1. We found that studies performed with this technique were of high quality and without loss of diagnostic information. All studies were considered technically diagnostic in this series, despite the mean tube current being only 52.2 ma. However, increased noise was the primary reason that 35% of the examinations were graded as diagnostic rather than excellent in quality. One potential risk of performing CT without sedation in young children is that of motion artifact caused by lack of cooperation. The need for cooperation is particularly important when IV contrast material is administered because of the limitations of scanning only during the optimal phase of the contrast bolus. Despite the low frequency of sedation in our series, only two studies one of which was in a sedated patient showed motion artifact, and this was seen on only a few images. Both of these examinations were considered diagnostic, and neither was compromised by the motion artifact. We found no statistically B 1526 AJR:179, December 2002

5 CT of Extrinsic Airway Compression in Children significant difference between the sedated and nonsedated groups in the percentage of excellent-quality examinations. Postprocessing techniques under investigation with CT include volume-rendered angiography and the reformatting of images in the sagittal or coronal planes [16, 17]. In our study, multiplanar or volume-rendered reconstructions were performed in 16 (30%) studies and were found to be diagnostically helpful in four of these cases. The standard axial images were generally found to provide the required diagnostic information regarding airway compression. Other investigations have shown reconstructed images to be of value in showing findings not seen on axial images [18]. Reconstructed and three-dimensional images are often of value in communicating findings to referring physicians. This study did not address the usefulness of such images for this purpose. The evaluation of the usefulness of the reconstructed images is limited beyond describing what occurs in our hospital practice. Postprocessing techniques were not used consistently: their use was dependent on which pediatric radiologist was monitoring the study. Some pediatric radiology staff created postprocessing images in all cases and others, in no cases. Other limitations of this study include its retrospective nature and small sample size. However, considering the low frequency of occurrence of airway compression in children, our sample size is relatively large. Additionally, the study was performed primarily to assess the technical feasibility of evaluating the pediatric airway with multidetector CT. The accuracy of low-tube-current multidetector CT for extrinsic compression was assessed by comparing the CT findings to results of endoscopic and surgical evaluation, which were available for most patients. No cases were found in which CT findings were shown to be inaccurate at surgery or endoscopy. Also, this study did not address the specific question of how much the tube-current setting can be reduced before increased noise renders an examination nondiagnostic for potential causes of airway compression. Our results show the feasibility of lowtube-current multidetector CT in the evaluation of pediatric patients suspected of having extrinsic airway compression. Multidetector CT offers an alternative to MR imaging, providing similar diagnostic information with a significantly reduced need for sedation. References 1. Donnelly LF, Strife JL, Bisset GS III. The spectrum of extrinsic lower airway compression in children: MR imaging. AJR 1997;168: Vogl T, Wilimzig C, Bilaniuk LT, et al. MR imaging in pediatric airway obstruction. J Comput Assist Tomogr 1990;14: Simoneaux SF, Bank ER, Webber JB, Parks WJ. MR imaging of the pediatric airway. Radio- Graphics 1995;15: Donnelly LF, Frush DP, Nelson RC. Multislice helical CT to facilitate combined CT of the neck, chest, abdomen, and pelvis in children. AJR 2000; 174: Donnelly LF, Emery KH, Brody AS, et al. Minimizing radiation dose for pediatric body applications of single-detector helical CT: strategies at a large children s hospital AJR 2001;176; Egelhoff JC, Ball WS Jr, Koch BL, Parks TD. Safety and efficacy of sedation in children using a structured sedation program. AJR 1997;168: American Academy of Pediatrics Committee on Drugs. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures. Pediatrics 1992;89: Berdon WE. Rings, slings, and other things: vascular compression of the infant trachea updated from the midcentury to the millennium. Radiology 2000;216: Malviya S, Voepel-Lewis T, Tait AR. Adverse events and risk factors associated with the sedation of children by nonanesthesiologists. Anesth Analg 1997;85: Vade A, Sukhani R, Dolenga M, Habisohn-Schuck C. Chloral hydrate sedation of children undergoing CT and MR imaging: safety as judged by American Academy of Pediatrics guidelines. AJR 1995;165: Kauffman RE, Banner W, Berlin CM. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures: Committee on Drugs, Section on Anesthesiology, American Academy of Pediatrics. Pediatrics 1992;89: Pruitt AW, Striker TW, Anyan WR, et al. Guidelines for elective use of conscious sedation, deep sedation, and general anesthesia in patients: Committee on Drugs, Section on Anesthesiology, American Academy of Pediatrics. Pediatrics 1985;76: Pappas JN, Donnelly LF, Frush DP. Reduced frequency of sedation of young children with multisection helical CT. Radiology 2000;215: Brenner DJ, Elliston CD, Hall EJ, Berdon WE. Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR 2001;176: Paterson A, Frush DP, Donnelly LF. Helical CT of the body: are settings adjusted for pediatric patients? AJR 2001;176: Hopkins KL, Patrick LE, Simoneaux SF, Bank ER, Parks WJ, Smith SS. Pediatric great vessel anomalies: initial clinical experience with spiral CT angiography. Radiology 1996;200: Glikeson RC, Ciancibello LM, Hejal RB, et al. Tracheobronchomalacia: dynamic airway evaluation with multidetector CT. AJR 2001;176: Sorantin E, Geiger B, Lindbichler F, Eber E, Schimpl G. CT-based virtual tracheobronchoscopy in children: comparison with axial CT and multiplanar reconstruction preliminary results. Pediatr Radiol 2002;32:8 15 AJR:179, December