Correspondence

Sir: I read the article published in your Journal regarding the modification to aid tracheostomy. I do agree with the authors on the fact that it is non original as from my clinical practice we do carry out a lot of tracheostomy operations on children at Bristol’s children’s hospital. We use a similar technique regarding applying stay sutures to the trachea but we still utilise rings 3 and 4, cutting through the thyroid isthmus using bipolar dissection which obviates the need for sewing the thyroid isthmus as well as providing adequate haemostasis.

We use 2/0 prolene for the stay sutures instead of silk as this reduces the chance of infection. I see no advantage in fashioning the tracheostomy in children through rings 4 and 5 as in the child there is the risk of puncturing the innominate vein. I believe that this technique may have a role in adults.

OMAR M AYOUB
St Michael’s Hospital and Bristol Children’s Hospital, Bristol, UK

Sir: I read Weight and Nicholson’s paper with interest. They are correct in assuming that the practice of using stay sutures to aid the tracheal incision and to facilitate recannulation in the event of tube dislodgement is well established. They could have found reference to it in most standard otolaryngology texts.¹

I strongly disagree with the author’s advocacy of low tracheal incision (5th tracheal ring) and using the stay sutures to “pull the trachea up and away from...the subclavian vein, particularly in children”. It is in the child that this method is to be particularly avoided. In children, such a low tracheal incision tends to retreat into the chest on neck flexion; this is (1) uncomfortable for the child and (2) difficult to recannulate should decannulation occur. Furthermore, the tip of the tracheostomy tube may enter the right main brochus.¹

Sir: The authors are correct in their assumption that the technique of tracheostomy that they describe is not new. The use of the stay sutures and a vertical incision through the wall of the trachea are standard practice in children, if not adults.

Low tracheostomies, particularly in patients with short thick necks, are not without problems. As the authors correctly point out, the trachea angles posteriorly in the lower neck and upper mediastinum necessitating operating at a deeper level than if the tracheostomy is placed higher. The anatomy of the inferior thyroid veins is variable and placing the tracheostomy below the thyroid isthmus does not necessarily ensure that these veins will not be encountered and produce troublesome bleeding if not adequately controlled. When the tracheostomy incision is made low in the trachea the curvature of the tracheostomy tube is often not optimal resulting in irritation and possible abrasion of the peri-stomal skin and the tracheal mucosa at the distal end of the tracheostomy tube. If the tube is placed very low in the neck there is potential for erosion of large vessels. Low tracheostomies in awkwardly shaped necks sometimes necessitate the use of commercially available flexible tracheostomy tubes.

Whilst the technique described by the authors may intuitively seem attractive it is not without its drawbacks, particularly for the long-term tracheostomee. The standard tracheostomy created higher in the trachea, with division of the thyroid isthmus, as necessary, has stood the test of time and is generally associated with minimal morbidity.

ROBIN L BLAIR
University of Dundee Otolaryngology, Ninewells Hospital and Medical School, Dundee, UK

Sir: I read with interest the above article by Mr Nicholson and Mr Weight. The authors described the technique of using stay sutures to aid bringing the trachea forward to the incision and opening up the longitudinal tracheostomy. They quite rightly did not claim originality of this technique as it is widely practised in the US and, certainly, in New York where I trained. This technique has been described in at least one standard surgical reference book,1 although the authors claimed to have found no reference in the medical literature. One further manoeuvre that can help stabilise the trachea and aid exposure during insertion of the tracheostomy tube is to use a tracheal hook to anchor the cricoid cartilage.^1,2 The authors also described placing stay sutures around the 5th ring and presumably performing the longitudinal tracheostomy at this level. I would like to point out that most surgical literature describe performing the tracheostomy at the level of the 2nd and 3rd rings,^1-4 as avoidance of placement of the tracheostomy tube below the 4th tracheal ring lowers the incidence of tracheo-innominate artery fistula.¹

Sir: We read with interest Weight and Nicholson’s article on the use of stay sutures on either side of a tracheostomy, a technique commonly practised by otolaryngologists, particularly in paediatric patients. In this latter age group the thyroid isthmus is divided and a vertical mid-line slit is made through the 2nd, 3rd and 4th tracheal rings rather than resecting any part of the trachea.

For this reason, and because the paediatric trachea is more compressible, of smaller diameter and situated more posteriorly than that in the adult, tracheostomy tube change is potentially hazardous. Accidental extubation or dislodgement can also have disastrous consequences, as the fistula tract is not formed in the first few days. Even established tracts can close remarkably quickly.

The use of stay sutures on either side of the midline prior to the tracheostomy helps to overcome these complications by maintaining the patency of the vertical tracheostomy. Rather than inserting these at the 5th tracheal ring (as described by Weight and Nicholson) where access is difficult with risks to the subclavian vein and pleura, we suggest placing these at the third and fourth rings.¹

We would like to highlight that use of the stay sutures itself can be hazardous particularly if they become crossed over or entangled during the tube change, thus, adding to the airway obstruction. Our practice involves labelling each stay suture ‘L’ or ‘R’ (left or right) and securing it to the anterior chest wall with tape rather than using a gauze roll.

Furthermore, the left suture is of longer length as an aid memoir in case the labels become detached. It is vital to explain to all inexperienced staff how to secure the airway in the event of decannulation, dislodgement or change of the tracheostomy tube by simply lifting the stay sutures upwards and laterally.

Other potential problems with the method are that the stay sutures may weaken the anterior tracheal wall and cut out. As silk has a tendency to become sodden, we prefer to use prolene 2/0 for the stay sutures.

TASS MALIK, D LUFF, DP MORRIS and J CURLEY
Preston Acute NHS Trust, Preston, UK

Sir: We wish to commend Weight and Nicholson on their submission.¹ The technique that they describe, of using two laterally placed sutures, inserted into the anterior wall of the trachea, is routinely performed in the Department of Oto-laryngology/Head and Neck Surgery in Sheffield. We would like to highlight that in the event of tracheostomy tube dislodgement during the early postoperative period, the doctor attending the patient needs to know which of the two stay sutures are on the left or right side of the trachea. After tracheostomy, these stay sutures may be crossed and, thus, their use will have the counterproductive effect of narrowing the tracheostomy.

We recommend, therefore, that the tracheostomy stay sutures be clearly labelled or that one of the sutures (e.g. the right) be consistently double-knotted at its free ends.

Sir: Thank you for asking me to reply to the various letters in relation to our publication “A modification to open aid tracheostomy”.

I think it is best if we make a general response to the letters as they make a number of common points. We would like to thank the authors of the various responses and contributions, which have been made to our short article describing a point of technique. These authors make a number of sensible suggestions but a number also raise objections to the use of incisions through the fifth tracheal ring, particularly in children.

Our experience of tracheostomy is virtually confined to ITU-based patients who require a temporary tracheostomy to help with airway management and weaning. We still feel that in this group, avoiding division if the thyroid isthmus is worthwhile whenever this can be achieved as it vastly simplifies the operation. We do, however, accept that division of the thyroid isthmus is occasionally necessary and may be particularly helpful with long-term tracheostomy patients. The other main point of the technique is that the vertical slit in the trachea should clearly be made by starting with division of the second tracheal ring and then dividing downwards until an adequate tracheotomy has been achieved. It is very unusual to require that this incision is taken beyond division of the fourth tracheal ring and we would agree division of the fifth tracheal ring does have potential dangers relating to the great vessels, particularly in children.

MICHAEL NICHOLSON
University of Leicester, Leicester Warwick Medical School, Leicester, UK

Sir: We read with interest the recent review by Steele and Webster¹ on altered cardiac function, and were pleased that this important condition has been highlighted in the Journal. As the general population ages and more patients now surviving an acute myocardial infarction, it is anticipated that heart failure will soon become an increasingly prevalent condition affecting the clinician, surgeon and economist. Nevertheless we feel that certain points need to be highlighted.

Heart failure occurs when the heart fails to provide blood sufficiently to all metabolising tissues in the body at normal filling pressures. It should be appreciated that Starling’s law (which is based on an experimental model!) is an over simplification of the multisystem nature of this disorder, which is characterised by abnormalities of cardiac, skeletal muscle and renal function, stimulation of the sympathetic nervous system, and a complex pattern of neurohormonal changes - all of which are activated in heart failure to increase cardiac output.²

These mechanisms are relevant because our current intervention for heart failure is largely directed at correcting these alterations.

Indeed, the aim of treatment of systolic heart failure in the acute stages is to preserve cardiac output and relieve symptoms. Once stable, therapy should be aimed at improvement of symptoms and outcome. In the surgical setting, the patient with underlying myocardial systolic dysfunction normally develops acute heart failure postoperatively or on the ITU, as a result of arrhythmias, perioperative myocardial ischaemia/infarction or iatrogenic fluid overload.

With regard to the treatment strategies for systolic heart failure highlighted by Steele and Webster¹ it should be remembered that the benefit of loop diuretics is evident even before diuresis is induced and this is likely to be due to a vasodilatory effect. All inotropes (perhaps with the exception of digoxin) increase mortality in heart failure, and dopamine has recently been shown to be no better than placebo to be ‘renoprotective’³ The calcium antagonists have variable effects in heart failure - verapamil and diltiazem may be detrimental in systolic dysfunction, the long-acting dihydropyridines (amiodipine, felodipine) are neutral in heart failure and short-acting nifedipine can even be harmful.²

It is particularly worth emphasising that blockade of the abnormal neurohormonal stimulation in heart failure saves lives.² For example, the renin-angiotension-aldosterone system can be inhibited by angiotension converting enzyme inhibitors and angiotension II receptor blockers, as well as spironolactone (an aldosterone antagonist).2 The abnormal sympathetic activation in heart failure can be inhibited by beta-blockers, which should now be considered in all patients with stable chronic heart failure.² Consideration should also be made towards the use of anti-coagulation, especially if cardiac thrombus or atrial fibrillation is present, and amiodarone, if ventricular arrhythmias occur. Unfortunately, none of these treatment modalities are fully validated in diastolic heart failure, and the strategies highlighted by Steele and Webster¹ are theoretical rather than proven in large randomised trials.

Heart failure is an ill-defined but important condition that clinicians and surgeons are now seeing more and more frequently. It remains a disease of poor prognosis, and patients with severe heart failure have a 6 month mortality >50% which is worse than some cancers. However, heart failure management has undergone tremendous changes in therapy as a result of basic and clinical research. These new treatments are improving patient outcome and quality of life and it is important that we must keep up to date with the changes.

Sir: We were pleased that Mr Makin and colleagues found our educational review interesting. Perhaps we should re-emphasise that the review was more focused on the patient seen in surgical clinics and operating theatres than those attending the cardiology clinic. Much of the review did indeed highlight the multisystem nature of the disease requiring a multidisciplinary approach to management. We are, however, naturally pleased that they would wish to re-emphasise this important point.

The letter suggests that we made no comment about some aspects of the pathophysiology and management of heart failure. The review contained references to the neurohumoral changes seen in heart failure and specific mention was also made of the common causes of heart failure in the post-operative patient, i.e. arrhythmias, ischaemia and fluid overload. We also made reference to the treatment of heart failure with diuretics and did indeed comment on alterations in compliance of the cardiovascular system occurring that resulted in the alleviation of symptoms. We specifically commented that the inotropes commonly used on intensive care patients have no place in the long-term management of the patient with heart failure. However, we would contend that in many cases their use would be life saving in the short-term. There was specifically no mention of the use of dopamine for renal dysfunction, since it is now well accepted by those who look after the critically ill that dopamine does not prevent renal failure and is probably detrimental in view of the occurrence of arrhythmias.

We agree that heart failure is a disease with a poor prognosis and we had hoped that our review would help to highlight this specific problem in those undergoing surgery.

NIGEL R WEBSTER
Anaesthesia and Intensive Care, Institute of Medical Sciences Aberdeen UK

Sir: We read with interest the article by Khaira et al (2001) on repair of incisional hernias in the February 2001 issue of the Journal.¹ Their technique exemplifies the combination of fascial and prosthetic repair with mesh placed beneath and sutured to the lateral rectus sheath leaves. Direct contact between the mesh and viscera is avoided by interposition of the sac and scar tissue, a useful manoeuvre. Recent publications in the literature indicate that mesh placed in direct contact with viscera may have deleterious consequences, such as mesh erosion and enterocutaneous fistula formation. As a result of the fibroblast response produced by the rough texture of the mesh, it becomes incorporated into the host’s granulation tissue.² Intestinal loops adhere to the prosthesis, which is the first stage in the development of biomaterial-related intestinal fistulae. Even if the initial results seem good, enterocutaneous fistulae may occur in the long-term, since the time taken for fistula to develop may be long - several years.^3,4 Khaira et al (2001) should be commended for an excellent idea and its successful incorporation into clinical practice.¹

Controversy exists as to the best position for the insertion of the prosthetic mesh patch in incisional hernia repair. Alternatives which can be used include the ‘on-lay’(prosthesis placed across defect and overlapping its margins, ‘in-lay’(mesh sutured to defect margins) and ‘sub-lay’ (pre-peritoneal meshplasty) mesh prosthesis. Khaira’s technique is not a true ‘on-lay’ graft because most of the mesh lies under the external oblique. This particular ‘on-lay’ graft (more correctly an inter-muscular graft) is simple to apply, tension-free, and without apparent side effects. It is associated with an active fibroblast response and elimination of recurrences as a problem.⁵ We incorporated a multi-layered fascial-peritoneal repair, as described by Lazaro da Silva, (1979) with the prosthetic mesh placed beneath the rectus muscles, a variant of the classical ‘sub-lay’ technique.⁶ Between 1992 and 1999, 40 patients with recurrent incisional hernias were subjected to repair using this technique, with excellent results and no recurrences, to-date.⁷ Our results and those of Khaira et al (2001) support the effectiveness of this important concept, namely the inter-muscular meshplasty.

Khara et al should be congratulated for their good work and excellent conclusions. We would like to thank you for bringing attention to their useful technique incorporating the principle of inter-muscular mesh incisional hernia repair.

KIRIEN T. KJOSSEV
Department of Surgery, Military Medical Academy, Sofia, Bulgaria

JULIAN E. LOSANOFF
Department of Surgery, University of Missouri-Columbia, Columbia, USA

Sir: We are most grateful for the interest shown by Drs Kjossev and Losanoff regarding our article and agree with them that this is a simple technique that is widely applicable and gives excellent results. The only real drawback to this technique is that a part of the mesh is covered directly by skin and, therefore, the incidence of seroma formation is high. Most seromas, however, are dealt with by simple aspiration. It would be interesting to know if stitching the mesh under the rectus muscles improves lymphatic drainage and, therefore, decreases the risk of seroma formation.

Facial nerve palsy following intra-oral surgery performed with local anaesthesia
2000; 45: 330-3

Sir: I read with interest Mr Cousin's article on facial nerve palsy.¹ I feel it is important to correct some points relating to this article. The facial nerve grading system described by House and Brackmann has 6 not 5 grades, and is very specific about abnormalities of facial movement, rather than ‘normal, slight, moderate, moderately severe, and severe’.³ Case 5 was described ‘as having a Grade 1 lower motor neurone lesion’, by the House Brackmann grading system, this is normal. From the discussion, two further points of importance arise. Firstly, the chorda tympani supplies taste to the anterior 2/3 of the tongue, sensation otherwise is provided by the lingual branch of the mandibular division of the trigeminal nerve. Secondly, the fore-head muscles are spared in an upper motor neurone but not in lower motor neurone lesion, as stated in this article.

Sir: We read with interest the case series of facial nerve palsies after dental procedures. However, we feel obliged to point out some errors in using the House-Brackmann grading system of lower motor neurone (LNM) facial nerve palsy. The House-Brackmann facial nerve palsy grading system, first described in 1985, has subsequently been adopted as the gold standard in assessing LMN facial nerve palsy of any cause.¹ The benefits of a universally accepted grading system are that results and research can be compared from different centres, and that, clinically, changes in facial nerve function can be documented and understood by different clinicians. Being subjective, even trained observers experience significant inter-observer variation.² Therefore, if it is to be used, it is absolutely essential that clinicians using the system fully understand it. The system uses six (not five) grades as detailed in Table 1.

Grade	Description	Characteristics
I	Normal	Normal facial function in all areas
II sis	Mild dysfunction	Gross: slight weakness noticeable on close inspection; may have very slight synkinesis
		At rest: Forehead-moderate-to-good function
		Eye-complete closure with minimum effort
		Mouth-Slight asymmetry
III	Moderate dysfunction	Gross: obvious but not disfiguring difference between the two sides; noticeable but not severe synkinesis, contracture, and/or hemifacial spasm
		At rest: normal symmetry and tone
		Motion: Forehead - slight-to-moderate movement
		Eye - complete closure with effort
		Mouth - slightly weak with maximum effort
IV	Moderately severe dysfunction	Gross: obvious weakness and/or disfiguring asymmetry
		At rest: normal symmetry and tone
		Motion: Forehead - none
		Eye - incomplete closure
		Mouth - asymmetry with maximum effort
V	Severe dysfunction	Gross: only barely perceptible motion
		At rest: asymmetry
		Motion: Forehead - none
		Eye - incomplete closure
		Mouth - slight movement
VI	Total paralysis	No movement

In Case 5, the patient is described as having grade I lower motor neurone lesion. This is a contradiction in terms as grade I is normal. In Case 6, only the buccal branch is affected and the House-Brackmann system is not applicable to individual facial nerve branch palsies.

A further error is the claim that forehead muscles are spared in LMN lesions because of crossover innervation at a higher level. It is upper motor neurone lesions that can spare the forehead for the same reason.

I am pleased that the article has generated much correspondence and debate.¹I would like to thank Messrs Coatesworth, Karagama and Homer for their helpful comments. In particular I am grateful for clarification of the House Brackmann grading system for the classification of disorders of facial movement, and for the comments on forehead musculature in upper and lower motor neurone lesions of the facial nerve.

GCS COUSIN
East Lancashire Maxillofacial Service, The Royal Infirmary, Blackburn, UK

CORRESPONDENCE