Category: Future of Thoracic Surgery

Helen E. Davies; Andrew Rosenstengel; Y.C. Gary Lee

 Curr Opin Pulm Med. 2011;17(4):247-254.

Abstract and Introduction

Abstract

Purpose of review Pleural disease is common. Traditionally, many patients were subjected to surgery for diagnosis and treatment. Most pleural surgical procedures have not been subjected to high-quality clinical appraisal and their use is based on anecdotal series with selection bias. The evidence (or the lack) of benefits of surgery in common pleural conditions is reviewed.
Recent findings Recent studies do not support the routine therapeutic use of surgery in patients with malignant pleural effusions, empyema or mesothelioma. Four randomized studies have failed to show significant benefits of thoracoscopic poudrage over bedside pleurodesis. Surgery as first-line therapy for empyema was studied in four randomized studies with mixed results and no consistent benefits. Cumulative evidence suggests that radical surgery in mesothelioma, especially extrapleural pneumonectomy, is not justified. Advances in imaging modalities and histopathological tools have minimized the need for surgery in the workup of pleural effusions. Complications associated with surgery are increasingly recognized.
Summary Surgery has associated perioperative risks and costs, and residual pain is not uncommon. Many conventional pleural surgeries have not been assessed in randomized studies. Pulmonologists should be aware of the evidence that supports surgical interventions, or the lack of it, in order to make informed clinical decisions and optimize patient care.

Introduction

Pleural diseases are common; over 1 500 000 patients develop a new pleural effusion annually in the USA alone.^[1] Pleural effusion can arise from more than 60 causes, and establishing the cause and effective treatment can be challenging.

Thoracic surgery traditionally plays a major role in the workup and management of pleural effusions, from pleural biopsies to pleurodesis and from empyema to pneumothorax. Various aggressive pleural surgeries have been developed over the years: from the description of Clagett procedure in 1963^[2] – a three-stage radical procedure with chest wall resection to create a permanent open window for pleural empyema – to modern day extrapleural pneumonectomy (EPP) for mesothelioma, which involves resection of lung, chest wall, hemidiaphragm, pericardium and regional lymph nodes. Most pleural surgical procedures have not been subjected to high-quality clinical appraisal (let alone randomized studies) and their use is based largely on anecdotal series often flawed with selection bias.

The aim of management of pleural diseases is to deliver the diagnosis and best management with the least invasive procedure(s), shortest hospitalization period and lowest procedural morbidity and cost. Realization of the lack of evidence for many pleural surgeries, and the growing documentation of their procedure-related complications, has prompted the pleural community to examine ‘conventionally accepted’ pleural surgical approaches using randomized trials. Not surprisingly, many (e.g. thoracoscopic poudrage) have failed to demonstrate any significant benefits. Advances in imaging techniques, histopathology methods and therapeutic protocols further contribute to a reduction in need for invasive surgeries. Worldwide, in recent decades, the role of surgical intervention for the diagnosis and management of pleural disease has diminished significantly.

Clinicians must be critically aware of the evidence (or lack of evidence) supporting a specific surgical intervention before subjecting their patient to an operation. Progress can only be made if clinicians continue to challenge the truthfulness of ‘conventional wisdom’ and work toward less invasive means to achieve better patient care.

In this review, we discuss the role of surgery in commonly encountered pleural diseases and highlight the deficit in evidence that supports many ‘accepted’ surgical interventions, and the advances in pleural research which suggest parity or superiority of noninvasive approaches.

Surgery for Diagnosis of Pleural Effusions

A significant shift in the choice of diagnostic procedure for undiagnosed pleural effusion has been seen in recent years. Open thoracotomy, once the gold standard, has given way to less invasive video-assisted thoracoscopic surgery (VATS). In turn, VATS is giving way to the less invasive pleuroscopy (or medical thoracoscopy). VATS requires general anesthesia and is performed usually through two to four portals of entry. Pleuroscopy is performed usually by pulmonologists under conscious sedation with a single or double port of entry, often as a day case.

In the UK, the number of centers offering pleuroscopy has jumped from 11 to 37 in the past decade, significantly reducing the need for VATS or open pleural biopsies.^[3] Flexi-rigid pleuroscopy further increases the ease of the procedure over traditional rigid thoracoscopy and is gaining popularity.

This march toward less invasive procedures is in part driven by the realization that surgery carries a risk of chronic complications. Furrer et al.^[4] reported that chronic intercostal neuralgia (persistent pain) occurred in up to 44% of patients at 6 months postthoracotomy. In another series (n = 56), 9% of patients suffered from chronic postthoracotomy pain severe enough to require daily analgesia, nerve blocks, acupuncture or specialist pain clinic visits.^[5] It is not surprising that a systematic review favored VATS over thoracotomy, reporting lower analgesia requirements and a shorter length of hospital stay. However, VATS is still associated with persistent pain or discomfort at the operation site in over a third of patients after 3–18 months.^[4]

No studies directly compare VATS with pleuroscopy, but several large case series have suggested similar diagnostic efficacy for malignancy. Pooled results from case series evaluating pleuroscopy show a diagnostic sensitivity for malignant pleural disease of 92.6% (95% confidence interval 91.0%–93.9%),^[6–25] comparable to those achieved following VATS pleural biopsy.^[26,27] Pleuroscopy is a well-tolerated, cost-effective procedure. Mortality rates are low (<0.01%) and, in a series of over 6000 cases, surgical intervention was never required for hemostasis.^[28] Pleuroscopy is preferable over VATS if initial fluid analyses were uninformative, especially in suspected cases of malignant or tuberculous pleural effusions.

Furthermore, technological improvements in diagnostic imaging modalities have reduced the need for thoracoscopic biopsies. Computed tomography with pleural phase enhancement provides closer definition of the pleural surfaces and circumferential, nodular or mediastinal thickening, and a parietal pleural thickness of more than 1 cm provides diagnostic specificities of 100%, 94%, 88% and 94%, respectively, for malignant disease.^[29] Similar results were recently demonstrated by Qureshi et al.^[30] using pleural ultrasound.

In patients with radiological evidence of pleural thickening, the diagnostic sensitivity of imaging-guided and thoracoscopically obtained pleural biopsy samples is comparable (approaching 90%).^[3]

Advances in laboratory tests and biomarkers for pleural diseases also significantly reduce the need for pleural tissue biopsies. In many endemic countries, adenosine deaminase is used in the diagnosis of tuberculous effusion especially in patients with a compatible clinical picture and a lymphocyte-predominant effusion, negating thoracoscopic biopsies.^[27,31] Other examples include flow cytometry for diagnosing lymphoma from pleural fluids, amylase for pancreatic effusions or ruptured esophagus and beta-2 transferrin for duropleural fistulae.

In patients with suspected mesothelioma, the use of a rapidly growing number of biomarkers has been proposed to aid the diagnosis through serum or pleural fluid analyses (reviewed elsewhere^[32,33]). Although none can substitute a histocytological diagnosis, a high mesothelin level in cases with suspicious cytology of mesothelioma can add confidence to diagnostic certainty and may obviate the need for surgery.^[34] In a study of 167 prospective patients presenting with undiagnosed pleural effusion, a negative mesothelin level together with negative pleural fluid cytology for malignancy yield a negative predictive value of 94%^[34] – highly comparable to the false negative rate for pleuroscopy in three large series.^[13,35,36] It is anticipated that within the next decade, these biomarkers will have an established place in the diagnostic algorithms for common pleural conditions, further minimizing the need for thoracoscopy.

Surgery for Pleural Infection

Pleural infection is a centuries’ old problem, but its incidence continues to rise despite better medical care and antimicrobials. The principle of therapy is control of sepsis (antibiotics) and drainage of the infected pleural fluid collection by thoracentesis, and if this fails, surgical evacuation.

Empyema is still considered in many centers as a ‘surgical’ disease, where surgeons will insert large bore chest tubes and have a low threshold of performing thoracoscopy for fluid evacuation if there are residual radiographic opacities. The conventional belief of the benefits of surgery stemmed from many anecdotal series, flawed by selection bias. The magnitude of that bias has recently been quantified in a retrospective series of 4424 empyema patients in the USA over 20 years.^[37] Empyema patients selected for surgery were significantly younger by almost 10 years (52.9 vs. 61.5 years, P < 0.001) and had a significantly lower comorbidity index (0.8 vs. 1.4, P < 0.001).^[37] VATS procedures often (up to 17%^[38]) require conversion to open thoracotomy, thus increasing postoperative morbidity (see above). Many aspects of these ‘conventional practices’ are now shown to be overaggressive and unnecessary. There are several factors to consider.

First, the majority of patients with pleural infection can be adequately treated with antibiotics and chest tube drainage, without needing surgery. In the Multicentre Intrapleural Sepsis Trial (MIST) (n = 454), only 18% (n = 74) failed the above approach and required surgical treatment.^[39•] [This is akin to pneumothorax management, where a 20% recurrence risk after the first episode does not warrant automatic surgery.^[40] Routinely, subjecting all empyema patients to surgery is, therefore, unnecessary.

Four randomized clinical trials (RCTs) have now compared first-line VATS with conservative treatment (antibiotics and chest tube drainage with/without fibrinolytics). No major advantage (e.g. on mortality) has been documented with early surgical approach in all the trials. Two RCTs of pediatric empyema, comparing primary VATS intervention with chest drain and intrapleural fibrinolytic, both showed no advantages of early surgery. On the contrary, Sonnappa et al.^[41] showed that surgery was more expensive ($11 379 vs. $9127) but did not alter outcome over conservative management in 60 children with pleural infection. Higher hospital charges were observed in the study by St Peter et al.^[42] and similarly no significant differences in length of stay, oxygen requirement, days until afebrile or analgesia needed (n = 36). The two trials on adult empyema were small (n = 19 and 70, respectively) and difficult to interpret. Clear criteria to guide the need for surgical decortication, following the initial treatment administered postrandomization, are lacking in both studies.^[43,44] Bilgin et al.^[43] and Wait et al.^[44] both randomized patients for immediate VATS or chest drain and antibiotics +/− intrapleural fibrinolytic. Neither study showed a major benefit other than shorter hospital stays (8.7 vs. 12.8 and 8.3 vs. 12.8 days, respectively). Hence, a recent Cochrane review concluded that further studies are needed to determine best practice.^[45]

Supplementing improvements in antimicrobial therapies, imaging guidance of chest tube drainage is now increasingly used in place of surgical evacuation of pleural collections. This practice has reduced the amount of patients subjected to surgery, though the exact magnitude of the reduction is difficult to quantify.

Intrapleural therapy to aid the drainage also can negate the need for surgical evacuation. A large randomized study (n = 454) and subsequent meta-analysis have shown no benefit from intrapleural streptokinase therapy alone.^[39•,46] However, the combined intrapleural use of tissue plasminogen activator (tPA) and deoxyribonuclease (DNase) to breakdown adhesions and thin pus has synergistic benefits in preclinical models.^[47,48] This has led to a factorial trial of intrapleural tPA and DNase in patients with pleural infection. Preliminary results from the MIST-2 study (presented at the British Thoracic Society 2009 scientific meeting^[49]) appear promising: tPA and DNase improved radiological clearance of pleural abnormalities and reduced hospital stay. Only 5% of patients treated with this combination required surgical debridement.

Second, surgical decortication postempyema is grossly overemployed. Many centers submit patients to surgical decortication because of residual radiologic changes, even when sepsis had subsided. This practice is not supported by current clinical practice guidelines, which recommend surgery only in patients with persistent sepsis and a residual pleural collection despite appropriate drainage and antimicrobial therapy.^[50] Longitudinal follow-up data from large clinical studies showed that residual pleural opacities will resolve with time, as the inflammatory changes settle.^[39•,51] This is akin to radiographic parenchymal changes in patients with pneumonia.

Third, conventional teaching advocates large bore chest tube drainage for empyema and, in many centers, large drains are inserted only by thoracic surgeons. Traditionally, the main arguments for large catheters have been a better drainage rate, especially in draining pus, and a lower blockage rate. However, no evidence-based data concur with this supposition.^[52•] The difference in drainage rate for pus is not significant once the size of internal diameter of the catheter reaches at least 8F or above (~12–14F external diameter). Rates of drain blockage in empyema, another conventional concern, are similar in published literature between large and small bore drains; and regular flushing of small bore catheters often overcomes the problem of blockage.^[53]

Empyema fails to drain most commonly because of multiple septations, a hurdle which large drains will not overcome; increasing numbers of studies now show that larger drain size does not increase efficacy, even in empyema. In their study, analysing data on 405 patients with empyema, Rahman et al.^[52•] showed no significant difference in mortality, need for subsequent thoracic surgery, length of hospital stay, lung function or radiographic resolution in patients with chest tubes of varying sizes (<10F, 10–14F, 15–20F or >20F).

The main drawback of the large bore catheters is pain secondary to the larger incision and subcutaneous/transpleural tract required, as reported in several series.^[52•,54] Others have shown higher rates of infection with large tubes.^[55,56]

Surgery for Malignant Pleural Effusions

As many as 100 000 patients in Europe develop a malignant pleural effusion from lung cancer alone^[57] and 150 000 cancer patients in USA have a malignant pleural effusion each year.^[58] Little evidence suggests thoracic surgery has a salient therapeutic role in malignant effusion management, even though it is often employed worldwide.

Pleurodesis is considered the best therapy wherever suitable and, in head-on comparisons, talc has been shown to be superior to bleomycin, tetracycline or doxycycline.^[59–65] The optimal route for delivery of talc is controversial.

Talc poudrage administered by VATS is traditionally thought to be more effective than bedside slurry instilled via a chest tube. However, talc induces pleural mesothelial damage with subsequent pleural inflammation and symphysis, rather than acting as a glue;^[66–69] therefore, the supposed even distribution which results from insufflation is not essential for successful pleurodesis. Radioactive isotope studies have shown that talc can distribute around the pleural cavity by respiratory motions even if administered as slurry.^[70]

All randomized trials to date have failed to show a benefit of thoracoscopic talc poudrage over bedside chemical pleurodesis; three recent studies have compared talc poudrage with talc slurry,^{[71•,73,74]} and one, by Mohsen et al.,^[72] with povidone iodine. These are outlined in Table 1.^{[71•,72–74]}

[ CLOSE WINDOW ]

Table 1. Recent trials comparing talc slurry and bedside chemical pleurodesis

Reference; study design	Patient number	Primary cancer (TP/TS)	Length of follow up	Outcome measures	Result
Dresler et al. [71•];	TP=251	Lung: 89/93	Until death	Recurrence rate at 30 days:	No significant difference
RCT	TS=250	Breast: 59/56		TP: 145/251	Similar success rates at 30 days (75%) and efficacy at 6 months (50%)
				TS: 126/250 (P=NS)
Yim et al. [74];	TP=28	Lung: 18/15	Until death	Recurrence rate:	No significant difference
RCT	TS=29	Breast: 6/9		TP: 1/28
		GIT/other: 4/5		TS:3/29
				(P=NS)
				Complications:
				TP:3/28
				TS:2/29
Terra et al. [73];	TP=30	Breast: 15/19	6 months	Postpleurodesis lung expansion	No significant difference in Clinical outcome complications or quality of life
RCT	TC=30	Lung: 11/6		Radiological recurrence
		Lymphoma: 2/1		Clinical recurrence (requiring intervention):
		Unknown: 1/1		TP: 5/30
		Other: 1/3		TS: 4/30
				(P=NS)
Mohsen et al. [72];	TP=22	All breast	4 years	Recurrence requiring intervention:	No significant difference
RCT	PI=20			TP: 2/22
				PI: 3/20

GIT, gastrointestinal; NS, not significant; PI, povidone iodine; RCT, randomized controlled trial; TP, talc poudrage; TS, talc slurry

The largest trial by Dresler et al.^[71•] showed that talc poudrage at thoracoscopy induced significantly more complications than talc slurry pleurodesis. Rates of pneumonia requiring antibiotics, respiratory failure, bronchopleural fistulae, requirement for blood transfusion, atelectasis requiring more than two bronchoscopies, dysrhythmia, deep vein thrombosis, pulmonary embolism and postoperative death rates were all increased in the talc poudrage compared with the bedside talc slurry group.^[71•] Success rates of both techniques were similar (~75%) at 30 days after procedure. Efficacy reduced with time to approximately 50% at 6 months and a suggestion of a trend toward talc slurry being more effective.^[71•]

Indwelling Pleural Catheters

One major recent advance has been the increased utility of indwelling pleural catheters (IPC). These may be inserted as a day-case procedure, with local anesthesia and conscious sedation, thus reducing hospital time and avoiding the risks inherent to a general anaesthetic. It is now the preferred treatment method for patients with an underlying trapped lung and those who fail initial pleurodesis.^[75] Extending the use of IPC as a first-line treatment for patients with malignant pleural effusion is the subject of randomized trials in Europe. Recent series suggest that bedside insertion of IPC by pulmonologists or interventional radiologists is as well tolerated as surgical placement in the operating rooms.^[76]

Surgery for Malignant Pleural Mesothelioma

Perhaps the most aggressive pleural surgery performed nowadays is EPP. EPP is usually part of trimodality treatment in combination with chemotherapy and hemithoracic radiotherapy. Little high-quality evidence supports its use.

Even in the most experienced centers and despite surgical advances, the perioperative mortality rate remains approximately 4%.^[77] Other centers have observed similar findings; e.g. Rice et al.^[78] had 8% mortality in 100 cases of EPP; Stewart et al.^[79] had 7% mortality in 74 patients and Hasani et al.^[80] had 11% mortality in a series of 18 patients. There is also significant associated morbidity: Sugarbaker et al.^[77] report a complication rate in excess of 60%, a finding echoed by Schipper et al.^[81] (who also report a 3-year survival rate of only 14%). Life-threatening complications affect 25% of patients, including surgical complications requiring re-exploration (7%), cardiac arrest/myocardial infarction (5%), prolonged intubation (8%), deep vein thrombosis and renal failure.^[77] Late mortality (days 30–180) is significant, killing as many patients as in the first 30 days in one report. Additional morbidity arises from the chemotherapy and radiotherapy arms of the trimodality regime.

Despite this unacceptable safety profile, the trimodality approach does not cure mesothelioma. Alarmingly, though not surprisingly, Weder et al.^[82] reported worsening of quality of life in patients who underwent EPP, especially in physical, psychological and activity scores for at least up to 6 months after surgery. Although improved long-term survival has been claimed, the data are almost certainly a result of selection bias.

The Mesothelioma and Radical Surgery (MARS) trial was designed to address the role of EPP as a component of trimodality treatment in malignant mesothelioma.^[83] Even in the 300 patients believed to be potentially suitable and referred, only 50 were ultimately eligible after screening and were randomized; further confirming that EPP, even if useful, is applicable only to a minority of patients and will not make an impact on the global burden of mesothelioma.^[84]

Increasing data confirmed that EPP has a worse outcome than less radical surgery, for example pleurectomy/decortication. Flores et al.^[85] showed in a large but nonrandomized series that patients who underwent pleurectomy had improved survival compared with those who underwent EPP. Nakas et al.^[86] reported significant improvements in pain and dyspnea with VATS pleurectomy/decortication (n = 67) compared with EPP (n = 112), with improved 30 day mortality (VATS group 7.1% vs. EPP 23%), reduced hospital stay (14.3 days vs. 36.6 days) and overall mean survival (14.0 months vs. 11.5 months) in patients aged more than 65 years.

The most striking data to show the lack of surgical benefits came from Flores et al.,^[87] who in a large retrospective series showed a median survival of 14.3 months in patients undergoing EPP (n = 208) compared with 15.8 months (n = 176) following pleurectomy/decortication. Both were only marginally better than the median survival for patients (n = 174) who underwent explorative thoracotomy and were found to have extensive and inoperable disease (12.7 months).

Mesothelioma is not a solitary tumor but spreads along serosal surfaces. Surgery is not likely therefore to provide cure, as has been the observation to date. Because of the lag time between exposure and disease onset, the patients are often elderly with significant comorbidity, and current data do not support aggressive operations.

Surgery for Chylothorax

Although dietary manipulation may reduce chyle flow, patients with refractory chylothoraces often require surgical ligation of the thoracic duct if this fails, necessitating either VATS or thoracotomy. Increasing reports suggest that percutaneous thoracic duct embolization using fluoroscopic guidance may be effective and can obviate the need for invasive surgery.^[88,89]

Surgery for Pneumothorax

The majority of pneumothoraces can be managed without surgery. Patients with small primary spontaneous pneumothoraces (PSP) or with no symptoms, regardless of the size of the pneumothorax, may be safely treated with observation alone. Guidelines recommend initial pleural aspiration for patients with PSP and significant symptoms, and that any patient with a secondary spontaneous pneumothorax (SSP) has an intercostal chest drain inserted.^[41]

No evidence exists on which to base timing of referral for surgical intervention in patients with an ongoing air leak. International guidelines recommend that an opinion is sought within 2–5 days; however, this timeline is largely arbitrary.^[41,90]

Several retrospective studies argue against early surgical treatment. One retrospective review (n = 115) reported spontaneous resolution rates of 74% and 100% for those with PSP at 7 and 15 days, respectively; and 61% and 79% (at 7 and 14 days, respectively) for patients with SSP. Only five patients required surgical intervention.^[91] Two further studies of PSP showed that only 37% had an air leak at presentation, resolving in two thirds of cases within 1 week without intervention.

Ferraro et al.^[92] compared conservative (including tube thoracostomy) to surgical intervention (apical resection with either pleurectomy of pleural abrasion) for 366 patients with 508 episodes of spontaneous pneumothorax (239 patients with PSP, 127 with SSP). No significant difference was noted between the two groups in terms of recurrence rates.

Other nonsurgical approaches under exploration include ambulatory management with chest tube and one-way valve, and pleuroscopy. For patients with SSP, who are more likely to have a prolonged air leak and less likely to tolerate surgical intervention, prolonged observation, intercostal catheter drainage and use of flutter valves may preclude the need for surgery. Medical thoracoscopy as an alternative to VATS has increasingly been used for the management of spontaneous pneumothorax. Tschopp et al.^[93] in a RCT compared the efficacy of VATS pleurodesis (via abrasion or talc poudrage) to poudrage via medical thoracoscopy, showing no difference in long-term recurrence rate (approximately 5%).

Conclusion

For centuries, different surgical procedures have been used for various pleural diseases, without any quality data to prove their benefits over conservative alternatives. Surgery has associated perioperative risks and costs; and residual pain is not uncommon. To date, the randomized studies on surgical approaches have not shown a significant advantage in the settings of pleural infection, malignant effusions and mesothelioma. Pulmonologists should be aware of the evidence that supports surgical interventions, or the lack of it, in order to make well-informed clinical decisions and optimize patient care.

Sidebar

Key Points

• The overall aim of medical practice is to diagnose and treat with the least invasive methods.
• There is a paucity of randomized evidence to support surgical intervention for many pleural diseases and physicians need to be aware of this in order to make well-informed clinical decisions to optimize patient care.
• Radical surgery, especially extrapleural pneumonectomy, is not justified for patients with mesothelioma.
• Advances in pleural research suggest parity or improved outcomes with less interventional approaches for patients with empyema.

Acknowledgements
Professor Y.C.G.L. receives research grants from the Western Australian Health Department (State Health Research Advisory Council), Sir Charles Gairdner Research Funds, Raine Medical Research Foundation and the Cancer Council of Western Australia.

Curr Opin Pulm Med. 2011;17(4):247-254. © 2011 Lippincott Williams & Wilkins