Pneumonectomy for non-small cell lung cancer: predictors of operative mortality and survival.

OBJECTIVE: The purpose of this study was to identify predictors of operative mortality and survival following pneumonectomy for non-small cell lung cancer (NSCLC). METHODS: All 206 patients having a pneumonectomy for NSCLC between 1991 and 1997 in our unit were prospectively studied. There were 162 males (79%) and 44 females (21%) with a mean age (+/- standard deviation) of 61+/-7.7 years (range 34-81 years). Squamous cell (75%) and adenocarcinoma (17.0%) were the predominant histological types. The possible impact of 29 parameters on operative mortality and survival was tested with univariate and multivariate analysis. The mean follow-up was 2.3+/-1.2 years, ranging between 0 and 6.8 years, and it was complete. RESULTS: Operative mortality was 6.8% (14 deaths). On multiple logistic regression older age (P=0.04) and the development post-operatively of bronchopleural fistula (BPF) (P=0.01) were independent predictors of operative mortality. The overall, Kaplan-Meier, 1-, 3- and 5-year survival (+/- standard error from the mean), inclusive of operative mortality, was 68+/-3.3, 42+/-4.1 and 35+/-4.5%. On Cox proportional hazards regression adenocarcinoma (P=0.006), the development of BPF (P=0.003), older age (P=0.03) and higher pathological stage (P=0.02) were independent adverse predictors of survival. CONCLUSION: Pneumonectomy for NSCLC carries a considerable, but acceptable, operative mortality and provides an important survival benefit. This study suggests that older age and BPF are major determinants of an unfavourable in-hospital outcome; older age, BPF, adenocarcinoma cell type and higher pathological stage significantly reduce the probability of a long-term survival.

Airway fire during tracheostomy: prevention strategies for surgeons and anaesthetists.

Airway fires are an uncommon but real and devastating complication of tracheostomy. One such fire in a 31-year-old man is described. Surgical fires are discussed, and 15 reported cases of tracheostomy fire are reviewed. A tracheostomy protocol, adopted by our department and designed to avoid this life-threatening complication, is described. Surgeons and anaesthetists involved in tracheostomy must understand the fire hazard and how to avoid it.

Surgical treatment of para-oesophageal hiatal hernia.

The development of laparoscopic antireflux surgery has stimulated interest in laparoscopic para-oesophageal hiatal hernia repair. This review of our practice over 10 years using a standard transthoracic technique was undertaken to establish the safety and effectiveness of the open technique to allow comparison. Sixty patients with para-oesophageal hiatal hernia were operated on between 1989 and 1999. There were 38 women and 22 men with a median age of 69.5 years. There were 47 elective and 13 emergency presentations. Operation consisted of a left thoracotomy, hernia reduction and crural repair. An antireflux procedure was added in selected patients. There were no deaths among the elective cases and one among the emergency cases. Median follow-up time was 19 months. There was one recurrence (1.5%). Seven patients (12%) required a single oesophagoscopy and dilatation up to 2 years postoperatively but have been asymptomatic since. Two patients (3%) developed symptomatic reflux which has been well controlled on proton-pump inhibitors. Transthoracic para-oesophageal hernia repair can be safely performed with minimal recurrence.

A tailored surgical approach for gastro-oesophageal reflux disease: the Nottingham experience.

OBJECTIVE: The objective was to assess the results which can be achieved by tailoring the anti-reflux procedure to the anatomical and functional situation of the patient with gastro-oesophageal reflux disease (GORD). PATIENTS AND METHODS: Two hundred and seventy six patients undergoing a primary tailored anti-reflux procedure between 1986 and 1996 were evaluated. An anti-reflux procedure was selected on the basis of the anatomical and functional findings assessed by means of barium video, endoscopy, manometry and prolonged pH monitoring. The operations performed were Nissen fundoplication (77), total fundoplication gastroplasty (TFG; 140) and Belsey Mark IV (BMIV; 59). The unit policy is for life-long follow-up. The symptoms at review were assessed and graded according to previously published criteria (Orringer MB, Skinner DB, Besley HR. Long-term results of the mark IV operation for hiatal hernia and analyses of recurrences and their treatment. J Thorac Cardiovasc Surg 1972;63:25-31). Patients with recurrent symptoms were fully re-investigated. RESULTS: Mean hospital stay was 8.2 days (5-32 days). There was one hospital death (0.36%). Mean follow-up was 6.7 years (range, 2.2-13.1 years). Overall excellent or good results were achieved in 247 (89.5%) patients (92.2% in Nissen, 90.7% in TFG and 83.1% in BMIV group, P=0.1). In patients without oesophagitis (n=72), the success rate was 93.1%, while for patients with grade IV oesophagitis (n=89) this was 87.6% (P=0.2). Kaplan-Meier freedom from recurrent or new, operation-induced, symptoms at 10 years was 88.1% (89.5% in Nissen, 87.4% in TFG and 73.8% in BMIV groups, P=0.08). CONCLUSIONS: These data suggest that where the appropriate anti-reflux procedure is selected, surgery can achieve satisfactory mid- and long-term success rates across the spectrum of GORD. When oesophageal shortening is evident, or merely suspected, we favour a TFG. In the presence of impaired motility and no evidence of oesophageal shortening, a BMIV is the preferred approach. The Nissen procedure is used for uncomplicated cases.

An atypical presentation of spontaneous pneumomediastinum.

Spontaneous pneumomediastinum is a rare condition and a most uncommon complication of sporting activity. We describe a case of spontaneous pneumomediastinum in a 17-year-old boy while playing football with no history of blunt trauma to the chest. The patient presented with symptoms and signs suggestive of an oesophageal perforation. The importance of correct investigations and subsequent management are discussed.

Comparison of long-term results of total fundoplication gastroplasty and Belsey Mark IV antireflux operations in relation to the severity of oesophagitis.

OBJECTIVE: Belsey Mark IV (BM IV) and total fundoplication gastroplasty (TFG) were the standard anti-reflux operations in two consecutive periods in Nottingham City Hospital Thoracic Surgery Unit. The aim of this study was to compare the long-term results obtained by these two procedures emphasizing their relation to the severity of the oesophageal mucosal damage. METHODS: Ninety patients (50 females and 40 males with a mean age of 57 years) who had a BM IV operation between 1976 and 1983 and 86 patients (46 females and 40 males, with a mean age of 56.5 years) undergoing a TFG procedure between 1983 and 1986 were evaluated. All patients were assessed preoperatively by means of clinical history, barium meal and endoscopy. In addition, 72 of the patients having a TFG had prolonged pH monitoring and manometric studies. The unit policy is for life-long follow-up. The symptoms at review were assessed and graded according to the criteria published by Orringer et al. (Orringer MB, Skinner DB, Belsey RHR. Long-term results of the Mark IV operation for hiatal hernia and analyses of recurrences and their treatment. J Thorac Cardiovasc Surg 1972;63:25-33). RESULTS: In the BM IV group there was one post-operative death (1.1%). The median follow-up was 11 years (range 3-18 years). Overall good results were achieved in 64 patients (71.9%). In patients without oesophagitis (n = 24) the success rate was 91.7% while for grades I (n = 17), II-III (n = 36) and IV (n = 12) oesophagitis this was 76.5, 66.7 and 41.7%, respectively (P = 0.01). The actuarial success rate at 10 through to 18 years was 71.0%. In the TFG group there was no postoperative death. The median follow-up was 10 years (range 2-14 years). Overall good results were achieved in 78 patients (90.7%). In the absence of oesophagitis (n = 10) the success rate was 90.0% and for grades I (n = 12), II-III (n = 26) and IV (n = 38) oesophagitis this was 91.6, 92.3 and 89.4%, respectively. The actuarial success rate at 10 through to 14 years was 90.3%. The differences in the overall success rate (P = 0.002), the success rates forgrades II-III (P = 0.02) and IV (P = 0.001) oesophagitis and the long-term actuarial success rates (P = 0.001) were significant. CONCLUSION: These data provide evidence on the superiority of the TFG against the BM IV in achieving long-term relief of reflux symptoms in the presence of severe oesophagitis. We believe that failure of BM IV in this setting is due to obvious or subtle oesophageal shortening.

Chylothorax following oesophagogastrectomy for malignant disease.

OBJECTIVE: To define the incidence, causes, management and impact of Chylothorax after oesophagogastrectomy for malignant disease in Nottingham Thoracic Surgery unit. PATIENTS AND METHODS: Retrospective analysis of 523 patients with cancer of the oesophagus or the gastro-oesophageal junction who underwent oesophageal resection between January 1987 and November 1997 in a single unit using similar techniques and uniform routine perioperative management. RESULTS: Chylothorax occurred in 21 patients (4.0%). There were 12 males and 9 females with a mean age of 64.7 years (SD 7.5). Age, sex, tumour site, length, histological type, depth of wall penetration, nodal status and type of operative approach were not significant predisposing factors on univariate and multivariate analysis. Seventeen patients were treated conservatively (four deaths, 23.5%) and four surgically (one death, 25.0%), effective control of the chylous leak being achieved in all four cases. Eleven patients with a chylous drainage of up to 2.2 l/day, diminishing within 1 week of conservative treatment had an uneventful recovery. However, a chylous drainage of more than 2.5 l/day in the remaining ten patients was associated with increased morbidity, hospital stay, operative mortality and the need for surgical intervention. In comparison with the remaining patients (n = 502), those who developed chylothorax (n = 21) had more respiratory complications (42.8%, P = 0.008), longer mean hospital stay (23.8 days, P = 0.004), higher operative mortality (23.1%, P = 0.004) and, unexpectedly, reduced 5 year survival rate (P < 0.0001). CONCLUSIONS: There appeared to be no clear predisposing factor in the development of a chylous leak other than the routine extensive dissection. Although definitive conclusions can not be drawn, where there is early reduction of the initial amount (in this series up to 2.2 l/day) of drainage, there may be a place for successful non-surgical management; in cases of high output chylothorax, persisting after a few days of conservative treatment, however, early re-operation and ligation of the thoracic duct, seems to be advisable.

Differential role of vasoactive prostanoids in porcine and human isolated pulmonary arteries in response to endothelium-dependent relaxants.

The pig is increasingly being used in medical research, both as a model of the human cardiovascular system, and as a possible source of organs for xenotransplantation. However, little is known about the comparative functions of the vascular endothelium between porcine and human arteries. We have therefore compared the effects of two endothelium-dependent vasorelaxants, acetylcholine (ACh) and the Ca2+-ATPase inhibitor, cyclopiazonic acid (CPA) on the porcine and human isolated pulmonary artery using isometric tension recording. ACh and CPA produced endothelium-dependent relaxations of both the human and porcine pulmonary arteries. In the porcine pulmonary artery, the cyclo-oxygenase inhibitor, flurbiprofen had no effect on relaxations to ACh (Emax: control 67.8+/-8.8% versus 72.4+/-9.5% (n=11)) or CPA (Emax: control 79.6+/-5.0% versus 94.0+/-10.6% (n=7)). The nitric oxide synthase inhibitor, L-NAME converted relaxations to both ACh and CPA into contractile responses (maximum response: ACh 30.0+/-11.1% (n = 10); CPA 80.4+/-26.2% (n = 8) of U46619-induced tone). These contractile responses in the presence of L-NAME were abolished by flurbiprofen. In the human pulmonary artery, L-NAME and flurbiprofen partly attenuated relaxations to ACh (Emax: control: 45.1+/-12.1%; flurbiprofen: 33.4+/-13.5%; L-NAME: 10.1+/-7.2%) and CPA (Emax: control: 78.1+/-5.5%; flurbiprofen: 69.6+/-7.2%; L-NAME 37.9+/-10.7% of U46619-induced tone). These responses were abolished by the combination of both inhibitors. We have demonstrated that while the release of nitric oxide is important in responses to endothelium-dependent vasorelaxants in both human and porcine pulmonary arteries, in the human arteries, there is an important role for vasorelaxant prostanoids whilst in the porcine arteries, vasoconstrictor prostanoids are released.

Surgery for esophageal cancer in elderly patients: the view from Nottingham.

OBJECTIVE: Our aim was to compare the outcome of esophageal resection for carcinoma in elderly patients (aged over 70 and over 80 years) with that of younger patients managed within a single specialist thoracic surgery unit. PATIENTS AND METHODS: Between January 1987 and November 1997, 523 patients underwent esophagectomy for carcinoma in the Nottingham City Hospital Thoracic Surgery Unit. The patients were divided into 3 groups by age: group I, under 70 years (n = 337); group II, 70 to 79 years (n = 150), and group III, 80 to 86 years (n = 36). These groups were compared with regard to preoperative medical status, operability and resectability, complications, operative mortality, and longterm survival. RESULTS: Patients in groups II (6.0%) and III (2.8%) had fewer preexisting respiratory problems than patients in group I (12.5%), and the patients in group III had fewer preexisting cardiovascular problems (16.7%) than patients in groups I (25.2%) and II (32.7 %). Although patients in group III were generally less likely to have operable lesions (64.3%), no significant differences in resectability rate were detected among the 3 groups (80.8%, 77.7%, and 80%). Elderly patients (groups II and III) had a higher incidence of overall (34% and 36.1%), respiratory (24.7% and 19.4%), and cardiovascular (7.3% and 11.1%) complications than those aged under 70 years (24.6%, 16.3%, and 2.1%, respectively). However, operative mortality (4.7%, 6.7%, and 5.6%) and 5-year survivals inclusive of operative mortality (25.1%, 21.2%, and 19.8%) were similar among the 3 groups. CONCLUSIONS: Accumulated experience in all aspects of perioperative management may account for a low hospital mortality in elderly patients despite a greater operative risk. The survival benefit is similar to that in the younger age groups, enforcing the view that esophagectomy within specialist thoracic units can be safely offered (in appropriately selected patients) with acceptable long-term survival in all age groups.

Management of benign oesophageal stricture by total fundoplication gastroplasty.

In the Department of Thoracic Surgery, City Hospital, Nottingham, we use total fundoplication gastroplasty routinely in the management of patients with benign peptic strictures who are unresponsive to medical treatment. This is an analysis of our results between 1983 and 1987. Fifty-six patients are included. There was no operative mortality. An overall good result was achieved in 83.9% of the patients. The results were better earlier in the disease (86.7% in grade II and 90.3% in grade I) than later (60% in grade III). We believe that conservative surgery should not be unduly delayed once medical management has failed. We now reserve resection for patients with fibrotic undilatable strictures and failures of conservative surgery.

Paraoesophageal hiatal hernias: when to operate.

Between 1979 and 1988, 29 cases of paraoesophageal hernia presented to one surgeon (F.D.S.). There were 23 women and six men and the mean(s.e.m.) age was 66.3(4.1) years. All were symptomatic and 13 hernias (45%) were complicated by gastric volvulus, haemorrhage or perforation. Ten (34%) had evidence of gastro-oesophageal reflux, suggesting a sliding component in these cases. Operation, mostly transthoracic, consisted of hernial reduction, crural repair and, if indicated, an antireflux procedure. There were three deaths. Two occurred as a result of spontaneous, intrathoracic perforation of the hernia. The third followed dilatation of a benign stricture 2 months after surgery. The only major complication was a pulmonary embolus with full recovery. The mean(s.e.m.) follow-up time was 47.6(7.8) months and there were no recurrences. This series confirms that symptomatic paraoesophageal hernias warrant early repair because of the frequency and severity of associated complications. Although debate continues as to whether this policy should be extended to asymptomatic paraoesophageal hernias, we suggest that this should be so.

Maintenance of serum calcium after total thyroparathyroidectomy.

We have studied calcium regulation in 11 consecutive patients undergoing radical surgery for upper aerodigestive tract malignancy. Eight patients received postoperative parenteral nutrition including calcium (19 mmol/day) and tri-iodothyronine (30 micrograms/day) supplementation. Three patients received enteral nutrition with calcium (70 mmol/day), 1.25 dihydroxycholecalciferol (1 microgram) and thyroxine (150 micrograms/day) via a nasogastric tube. Mean (SEM) corrected calcium fell from 2.42 (0.013) to 2.03 (0.036) mmol/l after 24 h (P less than 0.01). Replacement therapy generally maintained the serum calcium above 2.0 mmol/l. However, values were associated with only one episode of tetany. Phosphate increased from 1.10 (0.05) to 1.79 (0.11) mmol/l, 7-9 days postoperatively (P less than 0.001). Tubular calcium reabsorption fell and urinary calcium excretion rose, consistent with loss of parathyroid hormone (PTH) action on the distal nephron. However, the renal leak of calcium can be considerably reduced by concomitant salt depletion. This enhances proximal tubular sodium and calcium reabsorption thereby limiting calcium delivery to the distal nephron. This offsets the consequences of the loss of PTH which normally regulates distal calcium reabsorption.

Long-term results of the Belsey Mark IV antireflux operation in relation to the severity of esophagitis.

A retrospective analysis of the results of the Belsey Mark IV operation has been conducted, relating these to the degree of esophagitis present preoperatively. Analysis of 89 patients showed a 91.7% success rate in patients without esophagitis. Good results diminished steadily as the severity of esophagitis increased, giving 76.5%, 75%, 66.7%, and 50% success rates for first-, second-, third-, and fourth-degree esophagitis, respectively. We believe that shortening of the esophagus is an important factor in this. Shortening is obvious in third- and fourth-degree esophagitis but subtle in first- and second-degree esophagitis. We conclude that the Belsey operation is adequate for patients without esophagitis, but for patients with any degree of esophagitis more effective reflux control is needed.

Thoracic injuries sustained by the survivors of the M1 (Kegworth) aircraft accident. The Nottingham, Leicester, Derby, Belfast Study Group.

On 8 January 1989, a Boeing 737 carrying 126 passengers and crew crashed onto the M1 motorway killing 39 passengers. Of 87 initial survivors, 74 had major injuries making this an unusual accident as most aircraft crashes result in very few severely injured survivors. This prompted the setting up of a major study group, the Nottingham, Leicester, Derby, Belfast Study Group (NLDB) to examine in detail the accident and its aftermath. This paper is part of that work and is an initial survey of the thoracic injuries sustained by the 87 survivors. Twenty-three passengers sustained major chest trauma and all had major injury to other parts of the body. Five of these patients died within 12 h of admission. Various patterns of chest trauma emerged from this study, including an increasing incidence of rib fractures with age and a distinctive pattern of upper zone pulmonary contusion in younger patients.

Resection for carcinoma of the oesophagus.

One hundred and seventy-one patients with oesophageal carcinoma were seen in the 7-year period October 1976 to September 1983. One hundred and thirty-three cases (77.8%) were explored with a view to curative resection. This was accomplished in 98 cases (73.6% of explored patients, 57.3% of the entire group). Overall operative mortality was 10.2%. Of patients undergoing curative resection, 9% developed benign strictures which responded to 1-4 dilatations, while 14.8% developed local recurrence which marked the terminal event. After excluding operative deaths, 61.3% of patients survived for 1 year, 27% for 3 years and 17.3% for 5 years. Five-year survival for patients with squamous carcinoma was 36% while that for adenocarcinoma patients was only 3%. It is concluded that curative resection can be accomplished in a selected group of patients dealt with in a specialized unit with low operative mortality. This approach offers the patient good palliation and the best chance of prolonged survival.