Pediatric Tracheotomies: Changing Indications and Outcomes

Transcription

1 The Laryngoscope Lippincott Williams & Wilkins, Inc., Philadelphia 2 The American Laryngological, Rhinological a nd Otological Society, Inc. Pediatric Tracheotomies: Changing Indications and Outcomes J effrey D. Carron, MD; Craig S. Derkay, MD; Gerald L. Strope, MD; Jane E. Nosonchuk, RN, MSN; David H. Darrow, MD, DDS Objective/Hypothesis: To study the outcomes and complications associated with pediatric tracheotomy, as well as the changing trend in indications and out comes since 197. Study Design: Retrospective chart review at a major tertiary care children's hospital. Methods: On childr~n who underwent tracheotomy at Children's Hospital of the King's Daughters (Norfolk, VA) between 1988 and 1998, inpatient and outpatient records were reviewed. Of 218 tracheotomies, sufficient data were available on 24. Indications for tracheotomy were placed into the following six groups: craniofacial abnormalities (13%), upper airway ob struction (19o/o), prolonged intubation (26%), neuro logical impairment (27%), trauma (7%), and vocal fold paralysis (7%). Results: The average age at tracheotomy was 3.2 ±.6 years. Although the prolonged in tubation group was significantly younger than all others, the neurological impairment and trauma groups were significantly older. Decannulation was accomplished in 41 %. Time to decannulation was significantly higher in the neurological impairment and prolonged intubation groups, but was significantly shorter in the craniofacial group. Complications occurred in 44%. Overall mortality was 19o/o, with a 3.6% tracheotomy-related death rate. Comparison of our series to other published series of pediatric tracheotomies since 197 shows fewer being performed for airway infections and more for chronic diseases, with a corresponding increase in duration of tracheotomy and decreased decannulation rates. Conclusions: Tracheotomy is a procedure performed with relative frequency at tertiary care children's hospitals. While children receiving a tracheotomy have a high overall mortality, deaths are usually related to the underlying disease, not the tracheotomy itself. Key Words: Presented at the Meeting of the Southern Section of the American Laryngological, Rhinological and Otological Society, Inc., January 14, 2. From the Departments of Otola ryngology-head and Neck Surgery (J.D., c.s.o.,.11..) and Ped.iatric Pulmonology (G.L.S., J.E.N.), Eastern Virginia Medical School and Children's Hospital of the King's Daughters Norfolk, Virginia. ' Editor's Note: Thie Manuscript was accepted for publication March 3, 2. Send Correspondence to J effrey D. Carron, MD, Department of Otolaryngology- Head a nd Neck Surgery, Eastern Virginia Medical School 825 Fairfax Avenue, Suite 51, Norfolk, VA 2357, U.S.A. ' Laryngoscope 11: July 2 Pediatric tracheotomy, indications, outcomes; complications, fistula. Laryngoscope, 11: , 2 INTRODUCTION Tracheotomy is performed with relative frequency at tertiary care children's hospitals. In the pediatric population, the most common indications include prolonged v n tilator dependence, often resulting from the consequences of prematurity and bronchopulmonary dysplasia, and upper airway obstruction from craniofacial or structural abnormalities of the upper airway, or hypotonia stemming from neurological or neuromuscular disturbances. During the 197s, laryngotracheobronchitis and epiglottitis were common indications for tracheotomy in children before endotracheal intubation became a popular treatment. 1-3 Past published series reporting pediatric tracheotomy have likewise demonstrated hi gh decannulation rat s and low mortality. More recent studies have demonstrated a trend toward lower rates of decannulation and fewer tracheotomies overall, as fewer are performed for infectious indications. 4-8 Moreover, with improved home health and home ventilation programs, the out-ofhospital mortality h as decreased as well. 9 1 Innovations in the management ofsubglottic stenosis over the past 3 years have helped cure chjldren with this affiiction, and improved perioperative management has increased their decannulation rates Similarly, in children with craniofacial anomalies, more aggressive surgical management, such as mandibular advancement osteotomies, has led to earlier decannulation With improvements in home health care of the tracheotomized patient and less stigma associated with having a tracheotomy, more neurologically impaired children are being referred for this procedure. Indeed, where it was once considered aggressive management, tracheotomy is viewed today as conservative therapy for these patients in our institution. To study the changi ng trends in pediatric tracheotomy, a retrospective chart review was performed on all patients undergoing tracheotomy at our regional tertiary care children's hospital from 1988 to MATERIALS AND METHODS A complete medical record search on all children undergoing tracheotomy at Children's Hospital of the King's Daughters 199

2 (Norfolk, VA) during the yea rs 1988 to 1998, inclusive, was obtained, and the hospital charts r viewed. Outpatient charts from the Pediatric Pulmonology and Pediatric Otolaryngology offices were reviewed, in addition to inpatient charts. Specific paramet rs were recorded into an Excel database (Microsoft, Redmond, WA), includfog birth <lat, date of tracheotomy, indication, date of deca nnulation, wh ther a tracheocutaneous fistula (TCF) requiring closure resulted, and whether a laryngotracheal reconstruction was r quired to r move the tracheotomy tube. In addition, deaths and complications were r corded whenever found. With the exc ption of 12 patients, tracheotomies were perform ed und r th d.irection of one of two p diatric otolaryngology subspecialists. The stand ard tracheotomy procedure at our instituti on uses a vertical midline skin incision, defatting of the neck tissu s, v-rtical tracheal incision, and pl acement of traction sutur s into th trachea adjacent to the tube. Virtually all tubes are sutur cl in plac. All patients go to the pediatric intensive care unit after surgery. Preoperative or postoperative direct laryngoscopy and/or bronchoscopy is often performed, either to assess the airway anatomy and r suits of prolong cl intubation or to confirm tub position. Trach otomy tub s wer chang d 5 to 7 days after surgery, with the vast majority of th se changes performed at the bedside. The initial trach otomy tub change is p rform cl by the otolaryngology t am consi ting of at I ast two physicians and including ci th r an ttending or s nior-leve l resident. Par nts of children who wi ll v ntually be discharged with th ir tracheotomy und rgo an int nse education program that includes cardiopulmonary r suscitation training and instruction on tho care and changi ng of th trach otomy tub while supervised by a pulmonary sp cialty nurse. Long-t rm surveill anc typically includes fl exible bronchoscopy by tho p di atric pulmonologist at 6-month intervals or whenev r indi c t d by a change in clinical status, with rigid bronchoscopy and granuloma excision can i d out by the otolaryngologist when indicated. In addition, interval rigid or!lex.i ble bronchoscopy is often p rformed if the child is to be under anesthesia for a separ t procedur. In our practice, we do not commonly remove granulom ta when they are encountered. Instead, we recommend surgical excision if the child is approaching decannulation or if the granuloma is completely or nearly completely obstructing the airway and accid ntal d cannulation could be imm diately li fe-threat ning, or if use of a speaking valve is s v rely limi t d. RESULTS Two hundred eight en pediatric tracheotomies were perform d a t our institution from 1988 to Fourteen h ad no r cord or insufficient charting; thus 24 tracheotomi s in 197 patients wer available for r eview. For the purposes of data n alysis, pa tients were plac d into six groups: thos with anatomical or functional upper airway obstruction (UA ); thos with airway obstruction from a craniofacial syndrome (CF ); those with prolong d intubation or ventilator dependency (Pl); patients who ar e neurologically impaired (NJ); those who required tr ch eotomy as a result of trauma to the h ead or neck ; a nd childr n with bilateral true vocal fold paralysis (Table I). N urologically impair d patients with pharyng al coll pse and/or severe obstructive sleep apnea were plac d in th NI group; however, those with a distinct a natomical cause, such a s subglottic stenosis, were placed in th UAO group. Also, pa tients with chronic ventilator dep ndency as a result of a neuromuscular illness, such as W rdnig-hoffmann syndrome, were included in t he NI TABLE I. Primary Indications for Tracheotomy. Upper airway obstruction (N = 38) Subglottic stenosis (13) Tracheomalacia (8) Pharyngeal stenosis (2) Subglottic hemangioma (2) Tracheal stenosis (2) Other (11) Craniofacial syndromes (N = 27) Pierre-Robin sequence (5) Treacher-Collins disease (5) Beckwith-Wiedemann syndrome (3) Nager syndrome (2) CHARGE association (2) Pfeiffer syndrome (2) Other (5) Prolonged intubation (N = 53) Prematurity or BPD (37) Congenital heart disease (3) Pneumonia (2) Other (11) Neurologic impairment (N = 56) Cerebral palsy (18) Hypoxic encephalopathy (4) Werdnig-Hoffmann syndrome (4) Spina bifida (3) Chromosomal abnormality (3) Myasthenia gravis (2) Encephalopathy- unspecified (6) Other (16) Trauma (N = 15) Closed head Injury (12) Laryngeal fracture (1) Laryngeal or tracheal injury (1) Cervical spine fracture (1) Vocal fold paralysis (N = 15) Hydrocephalus (3) Leukodystrophy (1) Sandhoff disease (1) Neurofibromatosis (1) Mobius syndrome (1) Idiopathic or unspecified (7) group. The overall distribution of patients is shown in Figure 1. More than half (55%) of the tracheotomies were performed on children under 1 year of age (Fig. 2). The average age at the time of tracheotomy was 3.15 years, with a significantly higher average age in the NI and T groups (4.7 and 7.4 y, respectively, P <.5 by Student unpaired t test ). At 1.4 years, the PI group had a significantly lower average age at tracheotomy (P <.5) (Fig. 3). Patients who have not been seen by eith er the otolaryngology or pulmonology teams in the past 2 years (8%) and those who Laryngoscope 11 O: July 2 11 Carron et al. : Pediatric Tracheotomies

3 9 +...;...:;...--""'"...,_, """""'_...,.;,._...,.. a +...,;.,;..,...~_,..,~-"'"--~----~--~~~~ ~ ~ 6 "' s ~ 4 Overall UAO CF Pl NI T VFP p<.5, fp<.5 Fig. 1. Primary indications for tracheotomy by grouping. Pl = prolonged intubation; NI = neurologically impaired; T = trauma; VFP = vocal fold paralysis; UAO = upper airway obstruction; CF = craniofacial abnormality. relocated or were transferred to a long-term care facility in another city with the tracheotomy in place (7%) were not available for long-term follow -up. The remaining 85% have been followed for 6 to 116 months. Decannulation was successful in 41 % of all patients. The CF group had the highest rate (63%), while only 12.5% of NI patients were able to be decannulated (Fig. 4). The time to decannulation was significantly higher in the NI or PI groups (P <.5 and P <.5, respectively), while the CF patients had a significantly lower duration (P <.5) (Fig. 6). Laryngotracheoplasty was performed in 21 patients, 19 of whom are decannulated (9.4%). To avoid overestimation of decannulation rates, patients who were lost to follow-up or relocated were included in the denominators. The length of time for which the tracheotomy was required in patients who were decannulated is demonstrated in Figure 5. Of note, the NI group took almost twice as long as the average to reach decannulation (45.6 mo vs mo, P <.5). The most common complication was stomal granuloma, with 66 granulomata recorded in 42 patients. Other complications included TCF (n = 27), tracheal stenosis (n = 4), tube plugging with respiratory arrest (n = 4), accidental decannulation (n = 3), false passage creation (n = Fig. 3. Average age per group at the time of tracheotomy. Pl = prolonged intubation; NI = neurologically Impaired; T = trauma; VFP = vocal fold paralysis; UAO = upper airway obstruction; CF = craniofacial abnormality. 2), and stomal keloid formation (n = 2). Pneumomediastinum, suture abscess, innominate artery erosion, and sever bradycardi a under anesthesia were each seen once. Pneumothorax was encountered three times, with two of the pneumothoraces occurring after TCF closure. There were no intraoperative deaths and only two intraoperative complications (pneumothorax and severe bradycardia). The overall complication rate was 44%. Tracheocutaneous fistulae at our institution are generally closed when the tract remains open after 6 months of observation. A persistent fi stula resulted in 25 patients, for an overall fistula rate of 31 %. Twenty-four of these fistulae have been closed, and one patient in a long-term care facility who is minimally symptomatic is being observed. Two fi stulae were closed within a month of decannulation; one was closed in a pati nt whose family was relocating, and another for a child whose long-term care facility demanded it. Children who had a persistent TCF had a significantly longer mean duration of tracheotomy than those who did not develop one (35.7 mo vs mo, P =.2 by Student unpaired t test) (Figure 6). Persistent TCF resulted in 69.6% of patients who had a tracheotomy for more than 2 years, yet only 8.1% of those requiring a tracheotomy for less than 2 years required TCF closure. Vl ~ c Q) 12 1 ~ co 8 a.._ 6. L.. Q) 4.J:J E ::I 2. z < :!o l O Age in Years Fig. 2. Age at the time of tracheotomy. 7 "J""""~~..._~~...,...,._~... ~--~---~----~~~ 6 Overall UAO CF Pl NI T VFP Fig. 4. Decannulation rates by group. Pl = prolonged intubation; NI = neurologically impaired; T = trauma; VFP = vocal fold paralysis; UAO = upper airway obstruction; CF = craniofacial abnormality. Laryngoscope 11: July 2 111

4 :t22.2t Overall UAO CF Pl NI TR VFP 'p<.5, tp<.5 Fig. 5. Number of months before decannulation. Pl = prolonged intubation; Ni = neurologically impaired; T = trauma; VFP = vocal fold paralysis; UAO = upper airway obstruction ; CF = craniofacial abnormality i 3 25 :x zo Overall Fistula No Fistula p-.3 Fig. 6. Duration of tracheotomy versus presence of a persistent trach ocutaneous fistula % 27% Overall UAO CF Pl NI T VFP Fig. 7. Mortality rates by group. Pl = prolonged intubation; NI = neurologically impaired; T = trauma; VFP = vocal fold paralysis UAO = upper airway obstruction; CF = craniofacial abnormality. ' To demonstrate a shift in indications, tracheotomyrelated mortality, duration of tracheotomy, and decannulation, historical comparison was made to other series cove ring the years 197 to 1985 (Table II). In review of the presented series, one can observe a dramatic decrease in the number of procedures performed for airway infections (laryngotracheobronchitis or epiglottitis), with a resultant decrease in decannulation, and a concomitant increase in duration of tracheotomy. DISCUSSION Although tracheotomy is a relatively common procedure in the United States, the indications in children have changed ov r the past 3 years_. In the 197s, the most common indications for tracheotomy were infectious, including acute epiglottitis and laryngotracheobronchitis. 1 -.'.l However, since the late 197s and early 198s, nasotracheal intubation has virtually replac d tracheotomy for epiglottitis and laryngotrach obronchitis. 4 -s As a consequence of this changing trend, decannulation rates have tended to drop and average time to decannulation has risen (Table II). The decrease in tracheotomy-related deaths in children with long-term tracheotomies probably reflects improvements in caring for and monitoring these patients in the hospital, as well as advancements in parental teaching and home health Th complication rate (44%) and mortality rates (19% overall, 3.6% tracheotomy-related) in this series were comparable to those reported by other investigators a,s,1s Because of the retrospective data collection, the number of complications was probably underestimated. In addition pneumonia and tracheitis often ar not recorded as diag~ noses in tracheotomy pati ents, because the infection is usually low-grade and treated in an outpatient setting. A striking finding of this study was the ve ry poor decannulation rate observed with neurologically impaired children. The high mortality ra te of these patients was certainly a factor, but the majority of tracheotomies in this group were performed for a chronic, deteriorating condition. Therefore it is not surprising that only about one in eigh t of these children currently have their tracheotomy tube r moved. Placing indications into groups can be difficult, because terminology and classification can differ among authors and patients often have several diagnoses. 1-7.rn Although some Laryngoscope 11 O: July 2 112

5 Institution Years (N) Children's Hospital of the King's (N = 197) Da~hters, No olk University of California, San (N = 44) Francisco Red Cross War Memorial (N = 293) Hospital, Cape Town Hos~ital for Sick C ildren, Toronto (N = 319) University of Southern (N = 153)t California Children's Hospital of Philadelphia (N = 42) 'In patients who were successfully decannulated. tpatlents In the USC study were all < 3 years old. L TB = laryngotracheobronchltls. TABLE II. Comparison of Reported Series. 1 -:i.s.e Morta lit~ (Tracheotomy- elated) 19% (3.6%) 1.5% (%) 1% (2%) 24% (1%) 22% (3%) 28 % (8%) Percent With L TB Percent or Epiglottitis Decannulated Average Duration % mo % 35 2mo 55% 98 92% < 1 y 67% < 1 wk 14% mo 29% 83 2mo 45 % 72 6mo authors have used the specific terms ''laryngotracheobronchitis," "epiglottitis," and "diphtheria," others have simply used "infection" to classify these disorders. Some authors subclassify airway abnormality, laryngeal abnormality, craniofacial abnormality, subglottic stenosis, and other airway abnormalities separately, while others group them together. Prolonged intubation, ventilator dependence, neurological or neuromuscular disorder, and central nervous system abnormality can all be interr lated and are difficult to separate as well. Having many classifications can be helpful for the sake of specificity, but subclassifying the groups makes them much smaller, with meaningful comparison more difficult. The significance of segregating patien ts into a trauma group is the higher age at the time of tracheotomy. If this group were eliminated as a distinct entity, the patient with cervical spinal cord injury and the 12 patients with closed head injury could be included in the NI group, and those with airway trauma could be placed in the UAO group, without significantly changing the respective group data. CONCLUSION Neurologically impaired children requiring tracheotomy tend to be older at the time of tracheotomy, have a higher overall mortality, are less likely to have decannulation, and take longer before decannulation than other children with tracheotomies. Children who receive a tracheotomy because of prolonged intubation or ventilator dependency tend to be younger at the time of the tracheotomy and take longer before decannulation, with a relatively high overall mortality. Children requiring a tracheotomy after a trauma tend to be older when compared with all others. The most common trauma indication in our series was closed head injury. Children requiring tracheotomy for airway obstruction from a craniofacial abnormality h ave a high decannulation rate and a significantly lower time to Laryngoscope 11 O: July 2 decannulation when compared with other children with tracheotomies. Children who have decannulation within 2 years are unlikely to have a persistent TCF, whereas those who have a tracheotomy for more than 2 years are much more likely to retain a persistent TCF. Although children receiving a tracheotomy have a relatively high overall mortality (19%), deaths are usually related to the underlying disease, not the tracheotomy itself. BIBLIOGRAPHY 1. Wetmore RF, Handler SD, Potsic WP. Pediatric tracheostomy: experience during the past decade. Ann Otol Rhinol Laryngol 1982;91: Line WS, Hawkins DB, Kahlstrom, MacLaughlin EF, Ensley JL. Tracheotomy in infants and young children: the changing perspective Laryngo'scope 1986;96: Prescott CA, Vanlierde MJ. Tracheostomy in children: the Red Cross War Memorial Children's Hospital experience Int J Pediatr Otorhinolaryn.gol 1989;17: Carter P, Benjamin B. Ten-year review of pediatric tracheotomy. Ann Otol Rhinol Laryngol 1983;92: Crysdale WS, Feldman RI, Naito K. Tracheotomies: a 1-year experience in 319 children. Ann Otol Rhino[ Laryngol 1988;97: Duncan BW, Howell Ll, delorimier AA, Adzick SN, Harrison MR. Tracheostomy in children with emphasis on home care. J Pediatr Su.rg 1992;27: Palmer PM, Dutton JM, McCulloch TM, Smith R.JH. Trends in the use of tracheotomy in the pediatric patient: the Iowa experience. Head Neck 1995;17: Arcand P, Granger J. Pediatric tracheostomies: changing trends. J Otolaryngol 1988;17: Senders CW, Muntz HR, Schweiss D. Physician SW'vey on the care of children with tracheotomy. Am J Otolaryngol 1991; 12: Newton L, Chambers H, Ruben R.J, et al. Home care of the pediatric patient with a tracheostomy. Ann Otol Rhinol Laryngol 1979;91:

6 11. olton RT. Th problem ofp diatric laryngotracheal st nosis: a clinical and xp rimental study on the efficacy of autogenous cartilaginous gra!ls placed between the v rtically divided halv s of the post rior lamina of the cricoid cartilag. Laryngoscope 1991;11(12 Pt 2 uppl 56): Yellon RF, Param swaran M, Brandom BW. D creasing morbidity fo llowing l ryngotracheal r construction in chi ldren. Int J Pediatr Otorhinolarynf{ol 1997;41: ray SD, Mill r RP, Myer M, olton RT. Adjunctive rncasur s for succ ssfu l lnryngolracheal r construction. Ann Otol Rhinol Laryngol 1987;96: Nunn DR, Derkay CS, Darrow DH, Magee WP. Tracheotomy removal after early mandibular advancement in patients with pediatric craniofacial syndrome. Otolaryngol Head Neck Surg 1997;117:S187-S Williams JK, Maull D, Grayson BH, Longaker MT, McCarthy JG. Early decannulation with bi lateral mandibular distraction for tracheostomy-dependent patients. Plast Reconstr Surg 1999;13:48-57, Wetmore RF, Marsh RR, Thompson ME, Tom LWC. Pediatric tracheostomy: a changing procedure? Ann Otol Rhinol Laryngol 1999;18: Laryngoscope 11: July 2 114