CASE REPORT

Lobar torsion following thoraco-abdominal

oesophagogastrectomy

V. Felmine1 and M. Zuleika2

1 Specialist Trainee 4, Department of Anaesthesia, St George’s Hospital, Tooting, London, UK

2 Consultant in Anaesthesia and Intensive Care, Department of Anaesthesia and Intensive Care, Royal Surrey County

Hospital, Guildford, Surrey, UK

Summary

Following thoraco-abdominal oesophagogastrectomy for an adenocarcinoma of the lower oeso-

phagus, an 81-year-old female with no pre-existing respiratory disease could not be weaned from

mechanical ventilation. Right upper and middle lobe torsion were found at thoracotomy on

the 14th postoperative day. Both lobes were resected. The patient was discharged from hospital

after several postoperative complications. Pulmonary torsion is a rare, potentially life-threatening

complication of thoraco-abdominal oesophagogastrectomy. Differentiation from the more com-

mon postoesophagectomy pulmonary complications can be difficult. Early post-thoracotomy lung

opacification, in the absence of the expected degree of hypoxaemia, should trigger a suspicion of

pulmonary torsion.

........................................................................................................

Correspondence to: Dr Vinita Felmine

E-mail: vfelmine@hotmail.com

Accepted: 6 April 2009

The development of pulmonary complications is the most

important contributor to morbidity and mortality after

oesophageal resection [1]. Pulmonary torsion is a rare

complication. Torsion is defined as the act or process of

twisting, turning or rotating about an axis [2]. Pulmonary

torsion can occur spontaneously, following trauma or

after thoracic surgery [3]. One or more lobes or an entire

lung may undergo torsion.

We describe a case of lobar torsion following a thoraco-

abdominal oesophagogastrectomy. It was only recognised

at thoracotomy after 14 postoperative days on the

intensive care unit. This case report aims to increase

awareness of pulmonary torsion as a potential complica-

tion of non-pulmonary thoracic surgery and to identify

issues in its diagnosis and management.

Case report

An 81-year-old female with cerebrovascular disease and

mild stable Alzheimer’s disease but no pre-existing

respiratory disease was diagnosed with an adenocarcinoma

of the oesophagus. A two phase subtotal oesophagectomy

with two field lymph node dissection was performed

through a right thoracotomy with laparoscopic gastric

mobilisation under general anaesthesia and a thoracic

epidural.

At induction of general anaesthesia, placement of a

double lumen endobronchial tube was difficult due to

the apparent small size of the left main bronchus on

bronchoscopy. A microlaryngoscopy tube (ID 5.0 mm)

was used to intubate the left main bronchus and enable

one lung ventilation. However, one-lung ventilation was

difficult during the procedure and the right lung had to

be re-inflated several times. The patient’s trachea was

extubated at the end of surgery.

Within an hour of extubation, the patient’s trachea was

re-intubated due to hypoventilation and desaturation.

The immediate postoperative chest radiograph (Fig. 1)

showed that all lung zones were aerated. The patient

again failed extubation on the first postoperative day.

The chest radiograph done after re-intubation showed a

normal left lung but homogenous opacification of the

right lung with no volume loss and no tracheal shift

(Fig. 2). Since tracheal extubation was not imminent

a percutaneous tracheostomy was performed. Blood-

stained secretions were noted both before and after the

Anaesthesia, 2009, 64, pages 1130–1133 doi:10.1111/j.1365-2044.2009.05988.x.....................................................................................................................................................................................................................

� 2009 The Authors

1130 Journal compilation � 2009 The Association of Anaesthetists of Great Britain and Ireland

tracheostomy but it was unclear if the source was the

oropharynx or the tracheobronchial tree.

All three intercostals drains were removed by day 7.

The patient continued to produce variable amounts of

blood-stained sputum. While both the neutrophil count

and C-reactive protein were persistently elevated, the

patient remained afebrile and microbiology reports on

sputum and pleural fluid were negative. Failure to wean

from respiratory support prompted several investigations.

Serial chest radiographs showed persistent opacification

of the right lung. Bronchoscopy identified oedema and

hyperaemia of the carina and copious tenacious mucoid

secretions but no mention was made of partial or

complete obstruction of the right main bronchus. A

CT scan of the chest showed a thickened rim around

the right upper lobe and part of the right middle lobe

(Fig. 3). The loculated fluid collection in the right

hemithorax was suspected to be an empyema of the right

chest but only a small amount of dark transparent fluid

was drained.

As the patient continued to deteriorate (requiring

vasopressors), a right thoracotomy was done on the 14th

postoperative day. Since previous attempts at double lumen

endobronchial intubation had been unsuccessful, the

patient’s lungs were ventilated through a single lumen

tracheostomy. The bronchi and vascular pedicles of the

right upper and middle lobes were found to be torted to

180� with the middle lobe at the apex. The deep oblique

fissure extended down to the bronchus. This lack of

bridging tissue between the lobes suggests an increased risk

of torsion [4]. While the right lower lobe was normal in

appearance, the right upper and middle lobes were tense

and haemorrhagic. The affected lobes (upper and middle)

were untwisted and then resected. Extensive haemorrhagic

infarction of the upper and middle lobes of the right lung

was confirmed on microscopic examination. Complete

disruption of the normal pulmonary architecture and blood

vessel thrombosis were also noted. There was no evidence

of primary or metastatic malignancy in the resected tissue.

Bronchoscopy on the day following pulmonary resection

found copious blood-stained secretions in the left main

bronchus and both lobes of the left lung. A postoperative

chest radiograph showed widespread patchy consolidation

of the left lung (Fig. 4).

Figure 1 Immediate postoperative chest radiograph.

Figure 2 Postoperative day 1 chest radiograph showinghomogenous opacification of the right lung.

Figure 3 3 CT chest showing right middle (RML) and upper(RUL) lobes, bronchial cut-off (BC) and the right pulmonaryartery (RPA).

Anaesthesia, 2009, 64, pages 1130–1133 V. Felmine and M. Zuleika Æ Lobar torsion following thoraco-abdominal oesophagogastrectomy......................................................................................................................................................................................................................

� 2009 The Authors

Journal compilation � 2009 The Association of Anaesthetists of Great Britain and Ireland 1131

Postlobectomy, the patient’s multi-organ dysfunction

improved. During the 110 days spent on the intensive

care unit following lobectomy her recovery was compli-

cated by several lower respiratory tract infections. She

was discharged from hospital 145 days after oesophago-

gastrectomy.

Discussion

Pulmonary torsion is a rare complication of non-pulmo-

nary thoracic surgery. We have found only three reported

cases of pulmonary torsion following transthoracic

oesophageal surgery [5–7]. However, pulmonary torsion

is probably both under-diagnosed and under-reported [8].

It carries a high mortality and early diagnosis requires a

high index of suspicion. In this patient, repeated intra-

operative reinflation of the collapsed right lung and the

lack of bridging parenchyma between lobes may have

contributed to torsion.

Clinical findings can be related to the pathophysiology

of pulmonary torsion. The affected lung is neither

perfused nor ventilated. Hypoxaemia is not a prominent

finding as the ventilation and perfusion defects are

matched [9, 10]. Bronchial occlusion, if partial, can

result in overinflation of the distal lung but if complete

leads to accumulation of secretions. Vascular obstruction

leads to haemoptysis and pleural effusions. Venous

occlusion results in pulmonary congestion whereas

arterial occlusion can lead to infarction and gangrene

of lung. Ischaemic and necrotic lung parenchyma can

result in the systemic inflammatory response syndrome

(SIRS) [9].

Radiographic signs suggestive of pulmonary torsion are

rapid opacification of the ipsilateral lobe following

thoracic surgery (which may be mistaken for pleural

blood or effusion), a collapsed or consolidated lobe that

occupies an unusual position on a plain radiograph, a

change in position of an opacified lobe on serial chest

radiographs, hilar displacement in a direction inappropri-

ate for the atelectatic lobe and alteration in the normal

position and sweep of the pulmonary vasculature. Bron-

chial cutoff or distortion may be seen on a plain

radiograph but are best seen on CT scan [3]. Broncho-

scopy alone cannot exclude a diagnosis of pulmonary

torsion [9] since the bronchoscope may pass distal to a

partial obstruction as it did in our patient. Excessive

secretions, bronchial hyperaemia and oedema may be

seen.

The aims of treatment are to preserve viable lung and

to resect infarcted tissue. Options include resection of

affected lung (with or without untorsion) and untorsion

without resection. There have been reports of untorted

lung being salvaged [10, 11]. Untorting lung that is

ischaemic may release inflammatory factors and thrombi

into the circulation and infected secretions and ⁄ or blood

into the bronchial tree. A double lumen endobronchial

tube can protect the unaffected lung from contamination

and allow differential ventilation. Intra-operative Tren-

delenburg position and intravenous steroids have been

suggested to be beneficial [11]. Banki and Velmahos [9]

have proposed a diagnostic and therapeutic algorithm for

pulmonary torsion.

In our patient, an awareness of pulmonary torsion as

a complication of non-pulmonary thoracic surgery

would have resulted in earlier diagnosis and manage-

ment. Despite clinical and radiological findings sugges-

tive of pulmonary torsion, it was only recognised at

thoracotomy. Since the torted lobes were macroscop-

ically infarcted and unsalvageable, untorting them prior

to resection led to soiling of the left lung with pooled

secretions and blood. Once the diagnosis of pulmonary

torsion was made, the left lung should have been

protected with a right bronchial blocker or intubation

of the left main bronchus with a microlaryngoscopy

tube.

Knowledge of rare, potentially life-threatening com-

plications of common procedures can prevent morbidity

and mortality. Early post-thoracotomy lung opacification

in the absence of the expected degree of hypoxemia

should trigger a suspicion of pulmonary torsion.

Acknowledgements

Consent for publication was granted by the patient. We

would like to commend the exemplary dedication of the

surgical and intensive care teams which resulted in this

patient’s successful outcome.

Figure 4 Postpulmonary resection chest radiograph showingwidespread patchy consolidation of the left lung.

V. Felmine and M. Zuleika Æ Lobar torsion following thoraco-abdominal oesophagogastrectomy Anaesthesia, 2009, 64, pages 1130–1133......................................................................................................................................................................................................................

� 2009 The Authors

1132 Journal compilation � 2009 The Association of Anaesthetists of Great Britain and Ireland

References

1 Atkins BZ, D’Amico TA. Respiratory complications after

esophagectomy. Thoracic Surgery Clinics 2006; 16: 35–48.

2 Dorland’s Illustrated Medical Dictionary, 28th edn. Philadel-

phia: W.B. Saunders, 1994.

3 Felson B. Lung torsion: radiographic findings in nine cases.

Radiology 1987; 162: 631–8.

4 Moser ES, Proto AV. Lung torsion: case report and literature

review. Radiology 1987; 162: 639–43.

5 Fisher CF, Ammar T, Silvay G. Whole lung torsion after a

thoraco-abdominal esophagectomy. Anesthesiology 1997; 87:

162–4.

6 Chan MC, Scott JM, Mercer CD, Conlan AA. Intraopera-

tive whole-lung torsion producing pulmonary venous

infarction. The Annals of Thoracic Surgery 1994; 57: 1330–1.

7 Oddi MA, Traugott RC, Will RJ, Simmons RA, Treasure

RL, Schuchmann GF. Unrecognized intraoperative torsion

of the lung. Surgery 1981; 89: 390–3.

8 Wong PS, Goldstraw P. Pulmonary torsion: a questionnaire

survey and a survey of the literature. The Annals of Thoracic

Surgery 1992; 54: 286–8.

9 Banki F, Velmahos GC. Partial pulmonary torsion after

thoracotomy without pulmonary resection. The Journal

of Trauma 2005; 59: 476–9.

10 Kanaan S, Boswell WD, Hagen JA. Clinical and radio-

graphic signs lead to early detection of lobar torsion and

subsequent successful intervention. The Journal of Thoracic and

Cardiovascular Surgery 2006; 132: 720–1.

11 Moore RA, Forsythe MJ, Niguidula FN, McNicholas KW,

Clark DL. Anaesthesia for the patient with pulmonary lobar

torsion. Anesthesiology 1982; 57: 129–31.

Anaesthesia, 2009, 64, pages 1130–1133 V. Felmine and M. Zuleika Æ Lobar torsion following thoraco-abdominal oesophagogastrectomy......................................................................................................................................................................................................................

� 2009 The Authors

Journal compilation � 2009 The Association of Anaesthetists of Great Britain and Ireland 1133