Anaesthesia for massive retrosternal thyroidectomy in a tertiary referral centre

Transcription

1 British Journal of Anaesthesia 111 (4): (2013) Advance Access publication 19 May doi: /bja/aet151 Anaesthesia for massive retrosternal thyroidectomy in a tertiary referral centre G. A. Dempsey 1,3 *, J. A. Snell 1, R. Coathup 1 and T. M. Jones 2,3 1 Department of Anaesthesia and Critical Care and 2 Department of Head and Neck Surgery/Otolaryngology, University Hospital Aintree NHS Foundation Trust, Lower Lane, Liverpool L9 7AL, UK 3 Liverpool CR-UK Centre, Institute of Translational Medicine, 5th Floor Daulby Street, Liverpool L69 3GA, UK * Corresponding author: Department of Anaesthesia and Critical Care, University Hospital Aintree NHS Foundation Trust, Lower Lane, Liverpool L9 7AL, UK. ged.dempsey@aintree.nhs.uk Editor s key points Patients with a massive retrosternal goitre (mrsg) may present difficulties with airway management. Specific concerns include problems with tracheal intubation, mechanical ventilation, or postoperative tracheomalacia, but opinion on the optimal anaesthetic management is divided. A technique of i.v. induction, direct laryngoscopy, and tracheal intubation was successful in 18 patients with benign multi-nodular mrsgs. However, intubation and ventilation were impossible in one patient and an emergency tracheostomy was required. Background. Retrosternal goitre (RSG) is an uncommon problem encountered rarely by anaesthetists working outside specialized head and neck (H&N) surgical units. Traditional anaesthetic teaching warns of difficult airway management in these patients. The incidence and extent of these problems is unclear. Methods. We have performed a retrospective review of the anaesthetic management all patients with massive RSG (extending to the aortic arch or beyond) presenting for thyroidectomy at University Hospital Aintree from January 2007 to May Results. Five hundred and seventy-three patients underwent a thyroidectomy procedure at Aintree University Hospitals NHS Foundation Trust (AUH) between January 2007 and May Of these, 34 cases were documented as having a RSG. Review of each patient s preoperative computerized tomography imaging identified 19 patients with massive RSG. There was one case of failed intubation. All other patients underwent uneventful tracheal intubation via direct laryngoscopy. All glands were removed through the neck with no requirement to proceed to sternotomy. There were no instances of postoperative respiratory problems or tracheomalacia. Three patients suffered recurrent laryngeal nerve (RLN) injuries. Conclusion. When managed within a dedicated H&N operating theatre we have found a low incidence of difficult tracheal intubation, difficult mechanical ventilation nor postoperative respiratory difficulties in patients with massive RSG and mid-tracheal compression because of benign multi-nodular goitre. Surgical complications, however, are more frequent than those associated with cervical thyroidectomy with RLN injury and postoperative bleeding more likely. Keywords: airway management; anesthesia; general; thyroidectomy Accepted for publication: 18 February 2013 Retrosternal goitre (RSG) is an uncommon problem encountered rarely by anaesthetists working outside specialized head and neck (H&N) surgical units. While there are numerous reports in the literature concerning the surgical management of RSG 1 5 and ensuing problems 6 8 those considering anaesthetic management are frequently limited to isolated case reports 9 14 and small case series. 15 Although RSG is well described there is still much confusion as to the precise definition. In a review of 34 papers comprising 2426 patients, Huins and colleagues 5 found four definitions of RSG. These definitions included the presence of a gland which: (i) any part extends below the thoracic inlet with the patient in the surgical position, (ii) more than 50% lies within the thoracic cavity, (iii) extends to the level of the fourth thoracic vertebra on chest X-ray, and (iv) extends to the level of the aortic arch. It is clear from a clinical perspective that those patients with RSG to the level of the aortic arch and beyond are a very different cohort to those where the gland just extends into the thoracic inlet. Huins and colleagues, therefore, proposed a revised classification of RSG based on the anatomical extent of thyroid enlargement into the thoracic cavity. They divided RSG into three groups and suggested optimal surgical approaches when considering thyroidectomy (Table 1). In an attempt to better define anaesthetic morbidity related to RSG, we have performed a review of the anaesthetic & The Author [2013]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please journals.permissions@oup.com

2 Anaesthesia for massive retrosternal thyroidectomy BJA management all cases of massive RSG (mrsg), defined as Huins Grade 2 or 3, presenting for thyroidectomy at University Hospital Aintree from January 2007 to May Methods As this was a retrospective audit formal ethical approval was not sought, however, appropriate institutional agreements to carry out the project were secured. All thyroidectomies performed from January 2007 onwards were identified using the Operating Room Management Information System (ORMIS CSC Healthcare). These cases were then cross referenced with the hospital picture archiving and communications system to assess retrosternal extension using computerized tomography (CT) images taken before planned surgery. In an attempt to identify those patients most at risk of anaesthetic morbidity only those patients with mrsg were included. Data collected included patient characteristics (age, sex, and weight), presence of co-morbidities, and ASA grading. Additional information, which related to presenting symptoms, including the presence of preoperative dyspnoea, dysphagia, stridor, voice change or all was also collected. Reported preoperative examination findings including obesity (as defined by body mass index.30 kg m 22 ), venous congestion in the territory of the superior vena cava (SVC), anaesthetic airway assessment and, in particular, whether a difficult airway was anticipated were noted. Reported CT radiographic findings which were recorded as part of this study included the extent of thyroid gland descent into the thoracic cavity and the presence of tracheal narrowing, tracheal deviation or both from the midline. Anaesthetic data recorded included the method of induction of general anaesthesia (i.v. vs inhalation), tracheal intubation (awake vs post-induction), Cormack and Lehane grade of laryngoscopy, difficult or failed intubation, difficulty ventilating the lungs after tracheal intubation, and difficulties after tracheal extubation. Operative data included the operating theatre used (H&N vs general), procedure performed (total vs hemi-thyroidectomy), need for sternotomy, volume of intra-operative blood loss and the need for blood transfusion, identification of the recurrent laryngeal nerve (RLN), and identification of clinically evident tracheomalacia. Postoperative data recorded included the need for re-intubation (and reason), immediate postoperative bleeding requiring re-operation, postoperative critical Table 1 Huins and colleagues classification of RSGs and suggested surgical approach for thyroidectomy Grade Anatomical location Suggested surgical approach 1 Above aortic arch Cervical 2 Aortic arch to pericardium Manubriotomy 3 Below right atrium Sternotomy care admission (planned vs unplanned), hospital length of stay and identification of prolonged RLN palsy. Results Five hundred and seventy-three patients who had undergone a thyroidectomy procedure at Aintree University Hospitals NHS Foundation Trust (AUH) between January 2007 and May 2012 were identified. Of these, 34 cases were documented as having a RSG. Review of each patient s preoperative CT imaging identified 19 cases in which the thyroid gland descended retrosternally to the aortic arch or beyond. There were 20 attempted thyroidectomies in 19 patients. One procedure was abandoned because of a failed intubation, this patient subsequently underwent a successful excision (see below). The mean age was 65 years (range 35 93), there were 11 male and 8 female patients. Co-morbidities included hypertension (seven patients), chronic obstructive pulmonary disease (three patients), and obesity (eight patients). Major presenting clinical features are detailed in Table 2. Two patients had cervical venous distension suggestive of superior vena caval obstruction with the presence of thrombus in the internal jugular vein confirmed on CT scanning. One patient presented with an axillary vein thrombosis and unilateral Horner s syndrome and evidence of acute pulmonary emboli on CT pulmonary angiography. All patients had good mouth opening as indicated by Mallampati scoring, one patient had limited neck extension on clinical examination. The extent of mediastinal gland extension is presented in Table 2. All patients had evidence of tracheal narrowing on Table 2 Patient characteristic data and clinical features of patients successfully undergoing removal of RSG (n¼19) Mean age, yr (range) 65 (35 93) Percentage male 56 Mean tracheal deviation, mm (range) 27 (12 41) Mean tracheal diameter, mm (range) (% reduction) Clinical presentation 9.6 (5.8 15) (25 70) Asymptomatic 9 Shortness of breath 5 Shortness of breath when supine 4 Patient reported dysphagia 3 Clinically evident stridor 3 Degree of mediastinal extension To aortic arch 14 Beyond aortic arch but above tracheal 1 carina To tracheal carina 4 Histology Multi-nodular goitre 17 Multi-nodular goitre and thyroiditis 1 Follicular carcinoma 1 Weight of gland, g (range) 221 ( ) 595

3 BJA Dempsey et al. CT scan and tracheal deviation from the midline (Table 2). In all cases, maximal tracheal compression was between proximal and mid-trachea. Three patients presented with clinically evident stridor. Eighteen patients underwent i.v. induction of anaesthesia, one had an inhalation induction using sevoflurane and one had an awake fibreoptic intubation (AFOI). In the 18 patients undergoing i.v. induction of anaesthesia neuromuscular block was provided by atracurium in a dose of 0.5 mg kg 21 after confirmation of the ability to ventilate the lungs using a bag and mask technique. Two patients received succinylcholine 1 mg kg 21 followed by atracurium 0.5 mg kg 21. All patients undergoing i.v. induction had their tracheas intubated uneventfully (Cormack and Lehane classifications detailed in Table 3) as did the patient having the AFOI. The patient having inhalation induction of anaesthesia developed airway obstruction, was impossible to ventilate using a bag/ mask technique, rapidly progressed to a can t intubate, cannot ventilate scenario, and required an emergency tracheostomy. This was complicated by significant bleeding the intended thyroidectomy was therefore abandoned. The patient ultimately made a good recovery and had his tracheostomy tube removed. The patient re-presented 16 months later when he eventually underwent a successful thyroidectomy (patient having AFOI described above). Direct laryngoscopy after AFOI revealed a Grade II laryngoscopy. The two male patients and one female patient who presented with inspiratory stridor had tracheal diameters on CT scan of 9.5, 5.8, and 6 mm, respectively. Tracheal tubes used in these patients had an internal diameter of 8.0, 7.0, and 7.5 mm, respectively. Each of these patients underwent i.v. induction of anaesthesia after pre-oxygenation. The two male patients received atracurium 0.5 mg kg 21 after confirmation of the ability to ventilate the lungs using a bag and mask technique. The female patient received succinylcholine 1 mg kg 21 followed by atracurium 0.5 mg kg 21. After tracheal intubation, there were no instances of difficulty in instituting mechanical ventilation (suggestive of fixed tracheal obstruction). Surgical procedures undertaken were 13 hemi- and 6 total thyroidectomies with one procedure abandoned as detailed above. All operations were completed via the cervical approach with no requirement for manubriumectomy or sternotomy in any case. There was no requirement for intra-operative blood transfusion. Injury to the RLN occurred Table 3 Cormack and Lehane grade at laryngoscopy of all patients undergoing attempted removal of RSG (n¼20) Grade at laryngoscopy n I 15 II 3 III 1 Failed 1 in three cases (one neuropraxia and two nerve palsies). After thyroid gland removal, digital assessment of tracheal wall strength confirmed that there was no evidence of clinically relevant tracheomalacia in any patient. After operation no patient developed post-extubation stridor. There were three episodes of early postoperative bleeding that required re-intubation and further surgery to evacuate the haematoma and arrest the bleeding two of these patients had unplanned critical care admissions. Discussion Studies detailing the anaesthetic management of RSG are frequently limited to isolated case reports 9 14 and small case series. 15 The largest cohort detailing the anaesthetic management of such patients was reported by Bennett and colleagues 16 who considered the management of 1969 patients: 18 from their own institution with the remainder extracted from 12 published papers which they reviewed. Despite finding scant evidence of difficult intubation and postoperative tracheomalacia this study was not confined solely to patients with mrsg. Similarly Findlay and colleagues 6 report assessing the risk of tracheomalacia in 62 patients with significant tracheal compression was not confined to those with mrsg. Despite evidence in the surgical literature to demonstrate the increased morbidity and mortality associated with mrsg, there is little within those reported cases to guide the anaesthetist faced with a patient with a mrsg or the practical management thereof. Even published expert opinion often appears confusing and highly variable. 19 It is a commonly held view among anaesthetists that patients requiring retrosternal thyroidectomy frequently pose significant airway management problems (both pre-/post-tracheal intubation and post-extubation). In light of this possible increased morbidity and mortality associated with larger RSGs, we have confined this report to those patients with mrsg as defined above. Traditional anaesthetic teaching warns of difficult airway management in these patients. This is suggested to occur at all stages of the anaesthetic difficult bag and mask ventilation, difficult tracheal intubation, difficulties with positive pressure ventilation attributable to tracheal compression/deviation, and postoperative problems because of tracheomalacia. The incidence and extent of these problems is unclear. Even among experts, there is significant divergence of opinion as to appropriate management and likelihood of specific complications. Cook and colleagues 19 recently described a female patient with upper airway obstruction because of a thyroid goitre and sought international expert opinion regarding the management of such a patient. There were significant divergences in opinion with experts describing certain techniques advocated by others as unsafe and a lack of consensus on the likelihood of complications such as tracheomalacia. We have shown, in this small case series of mrsg, predominantly as a result of benign multi-nodular goitre, identified from our large tertiary referral 596

4 Anaesthesia for massive retrosternal thyroidectomy BJA practice, that such complications are unusual, having established no cases of clinically relevant tracheomalacia after surgery and only one instance of difficult tracheal intubation. When faced with mrsg, significant concerns for the anaesthetist are often related to induction of anaesthesia and tracheal intubation. Of the patients described herein, 18 underwent uneventful i.v. induction and tracheal intubation. All of these patients were managed primarily by a dedicated H&N team. The H&N operating theatres at AUH, as a result of being in the largest centralized H&N surgical unit in the UK, are in a separate dedicated theatre suite. The management of the difficult airway within these theatres is a frequent, almost daily, occurrence. All anaesthetic and support staff are well-versed in the management thereof, familiar with failed intubation drills and all necessary equipment is immediately available. In addition, there is the ready availability of senior H&N surgeons able to undertake advanced techniques such as rigid bronchoscopy when necessary. The patient who experienced the failed intubation described above was initially managed in a theatre suite not dedicated to the management of H&N cases. At the second procedure, the airway was secured by means of an AFOI and the operation proceeded uneventfully thereafter. While it is difficult to assess the true incidence of failed intubation in this small series, we would stress that there were no cases involving difficult airway management in those procedures undertaken within the H&N theatre suite. The low incidence of airway morbidity is supported by Findlay and colleagues 6 observation that in a study of 62 patients with significant tracheal compression attributable to thyroid enlargement 85% underwent standard i.v. induction and tracheal intubation and by Bennett and colleagues 16 who reported only six difficult intubations in their review of 1969 patients undergoing thyroid surgery for goitres of varying size. Consequently, we would suggest that for those patients with mrsg, because of benign goitre undergoing thyroidectomy, i.v. induction of anaesthesia and conventional direct laryngoscopy is a safe technique in appropriately experienced hands in a tertiary referral centre. However, we acknowledge and would suggest that in non-specialist centres an AFOI may be the most appropriate starting point for the less experienced operator. Review of airway management of cases of mrsg reveals a variety of techniques including AFOI, 9 awake direct laryngoscopy 13 and jet ventilation, 10 some of which would not usually be regarded as first-line procedures. The justification for many of these is the degree of tracheal compression produced by the RSG. As can be seen from the papers alluded to above, 616 there is little to support the contention that a RSG alone is a cause of difficult direct laryngoscopy. It would seem, therefore, that the presence of a RSG alone is unlikely to render tracheal intubation difficult. We would argue, therefore, that in such cases to use unfamiliar techniques, as described, is unnecessarily over complicating matters and potentially adding to the morbidity encountered. Indeed in cases of significant airway obstruction some authors would argue that the use of AFOI is debatable or even contraindicated, 20 a sentiment which is echoed in the recent report of the 4th National Audit Project of The Royal College of Anaesthetists and the Difficult Airway Society. 21 However, the presence of a RSG does pose problems should intubation difficulties arise. The ease with which surgical access to the trachea, either by needle cricothyroidotomy or formal tracheostomy, can be obtained when following a difficult airway algorithm are severely compromised in a patient population with significant goitres, tracheal compression, and deviation. We would, therefore, despite the concerns alluded to above by other authors, advocate that serious consideration be given to performing AFOI in those patients who have features other than their RSG that may pose difficulties in airway management. Despite a significant incidence of both tracheal narrowing (10 patients had tracheal diameters of,1 cm) and deviation in this patient population (Table 2), the tracheas of all patients were intubated with conventional sized tracheal tubes (internal diameters mm). The choice of tracheal tube size was predominantly based upon patient body habitus rather than on the degree of tracheal narrowing noted on CT images. Patients with significant tracheal narrowing were not managed significantly differently from those without. As the tracheal abnormalities described herein are almost always because of benign disease it is most likely that any narrowing present will be attributable to extrinsic compression rather than an infiltrative process. In such cases rigid fixed stenoses would be unusual. This fact is borne out by the observation that, despite significant narrowing in many cases, in this patient population, it was always possible to pass conventional sized tracheal tubes beyond the site of tracheal obstruction without difficulty. There were no incidences of difficult positive pressure ventilation after tracheal intubation. The phenomenon of extreme, post-induction, cardiorespiratory compromise in patients with mediastinal masses was first recognized in the 1970s. 22 Consequently, two of the experts consulted by Cook and colleagues 19 suggested they would use cardio-pulmonary bypass (CPB) as a Plan B option. However, reliance upon the use of CPB in this setting has been questioned. 23 None of the large surgical case series discussed herein report any instances of anaesthetic mortality. Sancho and colleagues 17 reported two deaths in a series of 35 patients of mrsg extending down to the tracheal carina. Neither of these was related to cardiovascular compromise under anaesthesia. We have been unable to find any reports in the literature of extreme cardiorespiratory instability because of mrsg in adult patients. Significant concerns also relate to the possible occurrence of post-thyroidectomy tracheomalacia. In Cook and colleagues paper expert opinion was divided as to the possible incidence of tracheomalacia in mrsg with some experts regarding its occurrence as highly likely with others regarding it as extremely rare. 19 In a study reviewing 34 papers totalling 2426 patients Huins and 597

5 BJA Dempsey et al. colleagues described an incidence of tracheomalacia of,1%. However, in those patients with RSG to the level of the aortic arch this was reported to be as high as 10%. Bennett and colleagues 16 review described six patients of tracheomalacia that were all reported within the same paper documenting 103 patients from Sudan. 7 Chauhan and Serpell 24 reported one case of tracheomalacia in a series of 199 RSGs not confined to mrsg. Findlay and colleagues 6 observed no cases of tracheomalacia in patients with significant tracheal compression because of thyroid goitres and concluded that post-thyroidectomy tracheomalacia remains almost mythical within modern thyroid surgery in the Western world. Given the small sample size, our incidences and complication rates could be questioned. However, our findings are similar to those published previously with mrsg with respect to airway management, 16 incidence of tracheomalacia, 56 postoperative bleeding and RLN injury. 18 The main limitations of the present study are its retrospective nature and small sample size. Given the relative rarity of mrsg this is difficult to address. Additionally, the majority of patients presented with benign multi-nodular goitres. Our findings and observations must, therefore, be interpreted with caution. It must be stressed that the relative ease of intubation and ability to easily pass the tracheal tube beyond regions of significant tracheal compression will not be applicable in other settings such as tracheal compression attributable to malignant thyroid disease or non-thyroid superior mediastinal masses. Because of the nature and relative rarity of mrsg meaningful studies into anaesthetic perioperative morbidity and mortality are likely to require multi-centre collaboration. In conclusion, perioperative anaesthetic management of mrsg is an uncommon problem outside dedicated H&N/ thyroid surgery units. Expert opinion on appropriate management remains somewhat polarized. We have presented details on the management of 19 patients which we believe illustrates that with the appropriate professionals, in the correct setting, the anaesthetic morbidity can be kept to a minimum. The incidences of difficult intubation, difficult mechanical ventilation, and tracheomalacia were low in this small series with mid-tracheal compression where the predominant pathology was benign multi-nodular goitre. Surgical complications, however, are more frequent than those associated with cervical thyroidectomy with RLN injury and postoperative bleeding more likely. Acknowledgements The authors would like to thank Drs B. Dupont, J. Rodrigues and Messrs S.R. Jackson, and R. Hardy for reviewing the manuscript before submission. Declaration of interest None declared. References 1 Raffaelli M, De Crea C, Ronti S, Bellantone R, Lombardi CP. Substernal goiters: incidence, surgical approach and complications in a tertiary referral center. Head Neck 2011; 33: de Perrot M, Fadel E, Mercier O, et al. Surgical management of mediastinal goiters: when is a sternotomy required? Thorac Cardiovasc Surg 2007; 55: Moran JC, Singer JA, Sardi A. Retrosternal goitre: a six year review. Am Surg 1998; 64: Ben Nun A, Soudack M, Best L-A. Retrosternal thyroid goiter: 15 years experience. Isr Med Assoc J 2006; 8: Huins CT, Georgalas C, Mehrzad H, Tolley NS. A new classification system for retrosternal goiter based on a systematic review of its complications and management. Int J Surg 2008; 6: Findlay JM, Sadler GP, Bridge H, Mihai R. Post-thyroidectomy tracheomalacia: minimal risk despite significant tracheal compression. Br J Anaesth 2011; 106: Rahim AA, Ahmed ME, Hassan MA. Respiratory complications after thyroidectomy and the need for tracheostomy in patients with a large goiter. Br J Surg 1999; 86: Shen WT, Kebebew E, Duh Q-Y, Clark OH. Predictors of airway complications after thyroidectomy for substernal goiter. Arch Surg 2004; 138: Younker D, Clark R, Coveler L. Fiberoptic endobronchial intubation for resection of an anterior mediastinal mass. Anesthesiology 1989; 70: Baraka A, Muallem M, Jamhoury M, Choueiry P. Jet ventilation in a case of tracheal obstruction secondary to a retrosternal goiter. Can J Anaesth 1993; 40: Nandwani N, Tidmarsh M, May AE. Retrosternal goitre: a cause of dyspnoea in pregnancy. Int J Obstet Anesth 1998; 1: Dave ST, Kamath SK, Shetty AN, Naik LD. Anaesthesia management for subtotal thyroidectomy in a case of multinodular goitre with retrosternal extension and superior vena caval syndrome. J Postgrad Med 2001; 47: Nakra D, Puri GD. Anaesthetic management of retrosternal goiter. J Anaesthesiol Clin Pharmacol 2005; 21: Patil S, Bhayani S. Airway obstruction caused by the systemic inflammatory syndrome associated with trauma and retrosternal goiter. Resuscitation 2009; 80: Radauceanu DS, Dunn J-OC, Lagattolla N, Farquhar-Thomson D. Temporary extracorporeal jugulosaphenous bypass for the perioperative management of patients with superior vena caval obstruction: a report of three cases. Anaesthesia 2009; 64: Bennett AMD, Hashmi SM, Premachandra DJ, Wright MM. The myth of tracheomalacia and difficult intubation in cases of retrosternal goitre. J Laryngol Otol 2004; 118: Sancho JJ, Kraimps JL, Sanchez-Blanco JM, et al. Increased mortality and morbidity associated with thyroidectomy for intrathoracic goiters reaching the carina tracheae. Arch Surg 2006; 141: Pieracci FM, Fahey TJ 3rd. Substernal thyroidectomy is associated with increased morbidity and mortality as compared with conventional cervical thyroidectomy. J Am Coll Surg 2007; 205: Cook TM, Morgan PJ, Hersch PE. Equal and opposite expert opinion. Airway obstruction caused by a retrosternal thyroid mass: management and prospective international expert opinion. Anaesthesia 2011; 66: Mason RA, Fielder CP. The obstructed airway in head and neck surgery. Anaesthesia 1999; 54: Popat M, Woodall N. Fibreoptic intubation: uses and omissions. In: Cook T, Woodall N, Frerk C, eds. 4th National Audit Project of 598

6 Anaesthesia for massive retrosternal thyroidectomy BJA the Royal College of Anaesthetists and the Difficult Airway Society. Major Complications of Airway Management in the United Kingdom. London: RCoA, 2011; Bitter D. Respiratory obstruction associated with induction of general anesthesia in a patient with mediastinal Hodgkin s disease. Anesth Analg 1975; 59: Slinger P, Karsli C. Management of the patient with a large anterior mediastinal mass: recurring myths. Curr Opin Anaesthesiol 2007; 20: Chauhan A, Serpell JW. Thyroidectomy is safe and effective for retrosternal goiter. ANZ J Surg 2006; 76: Handling editor: J. P. Thompson 599