Pneumothorax—Time for New Guidelines?

Rob Hallifax, DPhil, MRCP, BMCh; Julius P. Janssen, MD, PhD


Semin Respir Crit Care Med. 2019;40(3):314-322. 

In This Article



Patients commonly present with breathlessness and/or chest discomfort. The source of chest pain is not clear. Acute inflammation of the parietal pleura may play a role.[31] Alternatively, the pain could be the result of tearing of small adhesions, although one might expect some bleeding (hemopneumothorax), which is rare.

Tension pneumothorax, in which the hole in the visceral pleura forms a one-way valve, can lead to progressively increasing intrapleural pressure. In turn, this will lead to reduced venous return, hypotension, and hemodynamic compromise. These patients should be treated as an emergency with urgent decompression with a cannula anteriorly in the second intercostal space in the midclavicular line; treatment should not be delayed for radiological confirmation.[1] Thankfully, tension is thought to be a rare occurrence in PSP, typically occurring in the context of trauma or ventilated patients in intensive care units;[32] although true incidence in not clear.

In nontension pneumothoraces (Figure 2), patients with a small PSP who are clinically stable, conservative management (i.e., observation alone) with close radiological follow-up to ensure resolution may be appropriate. There is considerable controversy surrounding the treatment of minimally symptomatic patients with PSP, even in cases of a large PSP, and management practices vary.

Figure 2.

Chest radiography (CXR) image showing right-sided pneumothorax without mediastinal shift. The visceral pleura is clearly visible with an absence of lung markings distally.

Deciding who to Treat

The first choice to make is whether any intervention is required or whether the patient can be managed conservatively with observation.

For patients with PSP, the BTS guidelines suggest treatment is required in patients with large pneumothorax and/or symptomatic.[1] However, international guidelines and expert consensus statements vary.[1,2,33] The very definition of what constitutes a "large" pneumothorax is not consistent. The BTS definition is > 2 cm measured from the visible visceral pleura to lateral chest wall at the level of the hilum on CXR.[1] The American College of Chest Physicians consensus statement suggests using the distance from the lung apex to the ipsilateral thoracic cupola at the parietal surface, with large being ≥ 3 cm.[33] The Belgian Society of Pneumology use evidence of lung dehiscence over the whole length of the lateral chest wall to define a large pneumothorax.[34] A study of 49 pneumothorax episodes compared the classification of pneumothorax size across these three guidelines and found that there was agreement in only 47% of cases.[35] In addition to the lack of consensus on size definitions, there is an acknowledgement that large size does not always correspond to worse symptoms. The BTS guidelines include the caveat that "in some patients with a large pneumothorax but minimal symptoms conservative management may be appropriate." Therefore, the decision to treat or not largely lies in the hands of the treating clinician: patients with a small pneumothorax but severe symptoms may well require intervention; whereas a stable patient with a pneumothorax classified as large may be initially managed conservatively with observation only. It may be that the air leak was a transient event, which deflated the lung, but has already healed. These patients may be those with minimal symptoms that can be managed conservatively. Conversely, ongoing air leak will lead to progressive collapse or nonresolution of the pneumothorax, thereby requiring intervention to drain the pneumothorax and control air leak.

For patients with SSP, the current guidelines recommend admission to hospital for at least 24 hours and receive supplemental oxygen (unless suspected oxygen sensitive), and that "most" patients will require a chest drain.[1]

Time to Change the Guidelines?. A randomized controlled trial (RCT) of conservative versus active management of PSP has just completed recruitment in Australia and New Zealand, the results of which are expected soon.[36]

Needle Aspiration Versus Chest Drain

If a decision is made to intervene, the choice of initial management is needle aspiration (NA) or insertion of small-bore chest drain. NA (using 16–18G cannula) of up to 2.5 L of air is recommended as initial treatment. If aspiration is unsuccessful, then chest tube insertion is required. These recommendations are based upon a number of small RCTs showing equivalence of NA and chest drainage in terms of initial success rates: varying between 59 and 80%.[37–40] There were significant differences between the studies in terms of design and definition of "initial success" (Table 2).

In 2017, Carson-Chahhoud et al updated the Cochrane review,[45] this time including Harvey and Prescott,[37] Andrivet et al,[38] and Ayed et al[40] alongside Noppen et al,[39] with the addition of Parlak et al,[42] which also included traumatic pneumothoraces, as well as spontaneous. The authors concluded that immediate success rates were higher in the chest tube group than the NA group.[45] However, it should be noted that there was heterogeneity in the definitions of immediate success. Recurrence rates were no different between the groups. Hospital stay was shorter in the NA group (see below) as were the complication rates for NA (in a narrative synthesis) compared to the chest tube group.[45]

More recently, Thelle et al conducted another similar RCT enrolling both PSP (n = 79) and SSP (n = 48) patients.[43] They found NA to be more effective than chest tube drainage for initial treatment success (68.8% vs. 31.8%), although 37.5% (24/36) required a second aspiration.[43] NA remained significantly different when analyzing PSP (73.8% vs. 37.8%) and SSP separately (59.1% vs. 23.1%). This is the first study to specifically include sufficient numbers of patients with SSP to report the results separately. The authors suggest that the guidelines should be changed to treat SSP in the same way as PSP (i.e., with NA initially).[43] However, the low success rate of the chest tube drainage in all patient groups (31.8%) should be noted.

Time to Change the Guidelines?. The most recent Cochrane review concluded higher immediate success rates in the NA group compared to chest tube drainage, and may lead to change the current BTS guidelines of attempted NA first. Early failure rates and recurrence rates were not different between the groups, and hospital stay (if necessary at all) was shorter for the patients in the NA group. In addition, the recent study by Thelle et al again reports a high success rate for patients treated initially with NA.[43] Moreover, the success rate of NA in patients with SSP in this study was the same as that for PSP. The authors suggest NA could become first-line treatment in SSP, but this needs further validation in clinical trials.

Ambulatory Care

Given the benign course of many cases of SP (particularly PSP), the concept of ambulatory management has been considered, in order to reduce the inconvenience of patients and the economic cost being treated as an inpatient in hospital. This idea has been around for many decades, and a number of small studies show feasibility of outpatient management for PSP. A case series of 226 patients with PSP managed by observation or flutter valve concluded that outpatient management was "safe, efficient, and economical."[46] A randomized trial of 30 PSP patients (17 given "thoracic vent," 13 given standard chest drain) showed no significant difference in complications or reexpansion rates, but 70% of "vent" patients were managed as outpatients and required fewer analgesics, with patients in the control group (standard chest drain) hospitalized for 8 days.[47] Other studies have assessed efficacy of the one-way Heimlich valve attached to a standard chest drain in PSP, but there are no RCTs and two case-controlled studies showing differing results.[48,49] Another five nonrandomized, observational, or retrospective studies (totaling 195 patients) also describe high rates of success with outpatient management of PSP with one-way valves.[50–54] These data are well summarized in a 2013 systematic review of 18 studies using ambulatory management with Heimlich valve, reporting an overall success rate of 85.8% and successful outpatient management in 77.9% with "few complications." However, the evidence was of poor quality with a high risk of bias.[55]

Time to Change the Guidelines?. A multicenter study in the U.K., designed specifically to answer the question of efficacy and safety of ambulatory management, has recently completed recruitment and results are awaited (ISRCTN 79151659).[56]

Failure of Medical Management

"Failure" of medical management for patients with pneumothorax can be defined as a failure of chest drainage to reinflate the lung and/or prolonged air leak (PAL).

Use of Suction. Suction is application of negative pressure to the intrapleural space via the chest tube. Its routine use is not recommended by the current guidelines.[1] However, suction is advocated by some physicians in cases of nonresolution of pneumothorax (i.e., PAL and/or failure of lung reinflation). In this situation, applying high volume (up to 15–20 L/min) and low pressure suction (to prevent fast lung reexpansion and injury) is suggested.[1] The theory behind the use of suction is that it helps remove air in the pleural cavity a rate faster than it leaks through the visceral pleural breach, thus facilitating pneumothorax resolution.

Despite this guidance, there is no evidence for the use of suction. There have been two small RCTs,[57,58] neither of which showed a significant difference in success rates.[57,58] Furthermore, the application of suction is not without risk: specifically that of precipitating reexpansion pulmonary edema or exacerbating the hole in the visceral pleura, so its use should be carefully considered.

When to Define Treatment Failure?. There are some data to suggest that if one waits long enough, most PSP patients with PAL will eventually resolve. A study by Chee et al found that 100% of patients with PSP with PAL for > 7 days had resolved by 14 days.[59] Therefore, any decision to define failure has to be a balance between allowing time for the visceral pleura to heal (and PAL to cease spontaneously) and the practical limitations of patients remaining in hospital for up to 14 days.

Current BTS guidelines suggest that (if NA fails) drainage with chest tube should be attempted for at least 3 days, and a thoracic surgical opinion should be sought "early" at 3 to 5 days.[1] Because of the high failure rate of chest tube drainage after failed NA, this suggestion is subject to discussion.[60] Presumably, this surgical opinion recommendation is to make plans for possible intervention as they state that indications for surgery are PAL (despite 5–7 days of chest tube drainage) or failure of lung reexpansion.[1] The ideal timing for thoracic surgical intervention in cases of PAL and the cutoff point of 5 days is arbitrary. Alternatively, patients with PAL or failed NA can be sent for thoracoscopy and talc poudrage.