Occult traumatic pneumothorax: Is routine follow up chest X-ray necessary?

Highlights

• •Ideal follow up of occult pneumothoraces is not defined.
• •Occult pneumothoraces rarely require intervention.
• •Routine follow up chest X-rays are low yield.

Abstract

Background

Occult pneumothoraces (OPTX) are pneumothoraces that are not seen on chest X-ray (CXR) but visualized on computerized tomography (CT) scan. OPTX are frequently discovered during the initial trauma evaluation, there is a paucity of evidence based guidelines on how they should be further monitored. In this study we hypothesized that the practice of obtaining routine CXR for the surveillance of OPTX does not alter clinical management.

Methods

We retrospectively analyzed all adult (18 years or older) patients with a traumatic (blunt and penetrating) OPTX over a 7-year period (2013–2019) evaluated at an academic, Level 1, urban trauma center. Patient demographics mechanism of injury were abstracted. We examined subsequent radiographic and clinical evolution of OPTX as well as the need for intervention.

Results

363 patients were included in the series. Mean age was 47.8 years (range 18–98), 64.6% of the patients were male. Only 7 (1.9%) patients subsequently required an intervention for an OPTX. Neither age, sex, Injury severity score (ISS), or chest abbreviated injury score (AIS) (p = 0.072) were predictive of the need for intervention. Six of the seven patients that required intervention demonstrated clinical symptoms of pneumothorax progression.

Conclusions

In our experience the practice of obtaining routine chest X-rays for monitoring OPTX did not change clinical management. Obtaining a chest X-ray based on symptoms may be a more effective utilization of resources.

The use of computed tomography imaging has become ubiquitous in the work up of trauma patients. The increased sensitivity of CT imaging allows the ability of detecting small pneumothoraces that are not visible on plain film chest X-rays. These pneumothoraces, termed as occult pneumothoraces (OPTX) [1], may be diagnosed not only on chest but also on cervical spine and abdominal CT scans. It is estimated that half of small pneumothoraces are not detected on plain X-rays [2]. Some trauma registries report a 5% incidence of OPTX [1].

Diagram 1.

Diagram 1.

Early studies have demonstrated that OPTX can be safely observed without an intervention [3], [4], [5], [6]. Over the years additional studies have demonstrated that mechanically ventilated patients with OPTX can be also observed without thoracostomy drainage [7]. As our understanding of OPTX has evolved clinicians have adopted a conservative approach of observation and non-intervention of OPTX [5], [6], [7]. However, the literature does not prescribe specific observation protocols. In particular the frequency or interval for follow up imaging. Clinicians often obtain follow up chest X-rays to monitor for the potential progression of OPTX. These practices are based upon historical and clinical preferences and concerns that an OPTX could expand or evolve into a tension pneumothorax with detrimental outcomes. We hypothesized that the practice of obtaining routine CXR for the surveillance of OPTX does not alter clinical management among select trauma patients. In this study, we examined our trauma center's experience with the management of OPTX. We abstracted the radiographic evolution and clinical course of traumatic OPTX and the need for subsequent intervention among a subset of trauma patients.

Methods

We retrospectively analyzed patients with traumatic OPTX over a 7-year period (2013–2019) at an academic, urban, level 1 Trauma center. Patient data was extracted from the home institutions trauma registry. We included all trauma patients age 18 and older. OPTX were defined as pneumothoraces that were identified by attending radiology readings of computed tomography scan (CT scan) but could not be visualized on initial chest X-ray (CXR). Pneumothorax progression was defined as pneumothoraces deemed to have expanded on radiographic imaging by radiologist interpretation. Patients who were intubated on admission or had subcutaneous emphysema or hemothorax on the initial chest X-ray were excluded. We abstracted data for the number of follow up chest X-rays and the need for intervention to evacuate the pneumothorax. Pneumothorax symptoms included shortness of breath, crepitus, decreased breath sounds and or findings consistent with a tension pneumothorax physiology.

Statistical analysis was performed using IBM SPSS Statistics 28 (IBM Corporation, Somers, NY); statistical significance was assumed for p < 0.05. Median values are represented with 25–75% interquartile ranges. Mean values are presented with standard deviation. Binary logistic regression models were built to assess the impact of patient variables on need for intervention.

Results

Three hundred sixty-four patients met the inclusion criteria. The mean age was 47.8 years (range 18–98), 64% male. The median chest AIS was 3 (IQR 2–3). Patients received on average additional 1.27 chest X-rays after initial chest X-ray but there was a wide range from 0 to 7. Thirty-three (9.1%) patients demonstrated radiographic progression of OPTX on follow up X-rays but only 7/363 (1.9%) patients required an intervention (Table 1). On binary logistic regression, neither age (p = 0.456), sex (p = 0.634), Injury Severity Score (ISS) (p = 0.823) were predictive of need for intervention. Chest AIS correlated with need for intervention (p = 0.023) (Table 2). Six of seven patients that required intervention demonstrated clinical signs of pneumothorax progression (5 experienced dyspnea, 1 experienced oxygen desaturation). The seventh patient did not display symptoms but had a thoracostomy placed because of unexplained hypotension at a time of clinical decline. No deaths were attributed to the progression of an OPTX.

Table 1.

Patient characteristics.

Patient Characteristics	N = 363
Age (Years) +/_ standard deviation	47.8 +/_ 21.8
Gender% Male	N = 235 (64.5%)
Trauma Mechanism Blunt	N = 357 (N = 98.3%)
Physical Exam change among patients requiring intervention	N = 6/7 (85.7%)
Chest Abbreviated Injury Scale (AIS)	Mean 2.65, Median = 3, Mode = 3 (IQR = 2–3)
Follow-up chest X-rays	Mean 1.27, Median = 1, Mode = 1 (IQR 1–2) (Range = 0–7)
Length of Stay	5.4 days
Progression of PTX	N = 33 (9.1%)
Intervention	N = 7 (1.9%)

Table 2.

Patient characteristics and need for intervention. Binary logistic regression.

Age	p = 0.48
Sex	p = 0.63
Injury Severity Score (ISS)	p = 0.34
Chest abbreviated injury score (AIS)	p = 0.02

Discussion

OPTX is a common finding during the evaluation of trauma patients. OPTX has been extensively studied previously as questions arose on how to best manage this identity [1], [2], [3], [4], [5], [6], [7], [8]. OPTX rarely require intervention and thus may be safely managed with clinical observation. Our experience suggests routine follow up chest X-ray is not necessary. Over the years, the literature has supported a conservative approach of observation and avoiding chest tube placement [6,7]. However, the observation component of published guidelines is not well defined [6,9]. Currently, clinicians, at their discretion, often obtain follow up chest X-ray to monitor progression of OPTX. In this study, we examined the natural course of a selected group of patients with OPTX and the use of chest X-ray surveillance.

We decided to exclude patients who were intubated on arrival or soon after in the trauma bay. These patients were excluded from the current study because it would be difficult to decipher in a retrospective manner the reason for obtaining chest X-rays as it is a common practice for intubated patients to obtain X-rays for a multitude of reasons and thereby would cofound a retrospective analysis. Previous studies have demonstrated that intubated patients with OPTX can be observed and do not need immediate intervention [8], [9], [10], [11], [12]. Similarly, patients with an identified hemothorax were also excluded as the reason for subsequent X-rays would be also difficult to decipher as many of these patients underwent repeat imaging to monitor hemothorax progression. In addition, patients who had subcutaneous emphysema on plain films but no visible pneumothorax were not considered to be truly occult and were excluded from analysis.

In our series, a small minority of patients with OPTX required intervention. The number of chest X-rays were variable. Although a significant number of patients demonstrated radiographic progression of OPTX, most did not require a clinical intervention. Some of the reported radiographic progression may have been secondary to differences in technique as some patients underwent single view portable AP X-rays while some underwent two view PA films. This difference in modality may have contribute to some of the instances of OPTX being detected on repeat imaging and may not have represented an actual progression of the OPTX. Among those patients that required intervention, almost all displayed symptoms of pneumothorax progression. As described above, one patient did not display symptoms but had a thoracostomy placed empirically because of unexplained hypotension.

Our experience suggests that the progression of OPTX to the point of requiring intervention is rare overall and when required is evident by clinical symptoms. Our retrospective experience suggests that routine surveillance chest X-rays do not alter clinical management and are of limited clinical utility for OPTX. In our experience Chest AIS was the only factor that correlated with the need for intervention. This suggests that burden of chest trauma maybe a predictor for OPTX failing conservative management. Our experience suggests a selective approach to OPTX imaging maybe be more beneficial. Imaging based upon the development of symptoms would result in a better utilization of resources by reducing potential costs and avoiding unnecessary radiation exposure.

Our study has many limitations. The study is weakened by its retrospective design. Our exclusion criteria enabled us to extract data for isolated OPTX but excluded many trauma patients with OPTX including those critically ill intubated patients. Further multicenter prospective studies are needed to validate these findings before drawing firm conclusions.

Conclusions

In our experience OPTX rarely require intervention and thus may be safely managed with clinical observation. The routine use of follow-up chest X-rays is low yield and a practice of repeat imaging based upon symptoms may result in a better utilization of resources. Further prospective studies are needed to validate these findings.

Declaration of Competing Interest

The authors have no conflicts of interest to disclose.