Article, Pulmonology

Management of spontaneous pneumomediastinum: Are hospitalization and prophylactic antibiotics needed?

a b s t r a c t

Background: spontaneous pneumomediastinum is defined as the presence of Free air in the mediastinum without any apparent concomitant factors or disease. It is uncommon but usually benign and self-limiting. Generally, pa- tients with spontaneous pneumomediastinum are admitted to hospital, and occasionally, Prophylactic antibiotics are administered to prevent mediastinitis. The purpose of this study was to describe practices concerning the fea- sibility of outpatient treatment and follow-up of spontaneous pneumomediastinum and the necessity of prophy- lactic antibiotics.

Methods: We conducted this study in a single emergency medical center located in Kobe, Japan. We retrospec- tively evaluated patients with spontaneous pneumomediastinum from January 2007 to December 2014. Sponta- neous pneumomediastinum was defined as cases in which pneumomediastinum did not occur in the setting of positive pressure ventilation or severe trauma. All case records were carefully reviewed considering the demo- graphic data, symptoms, precipitating events, diagnostic workup performed, use of Prophylactic antibiotics, length of hospital stay, and complications.

Results: Thirty-four patients who satisfied the inclusion criteria were identified. Twenty-three patients (67.6%) were followed up on an outpatient basis, and 11 patients (32.4%) were admitted to the hospital, with a mean length of stay of 3.4 days. Prophylactic antibiotics were orally administered to 2 patients. None of the patients de- veloped complications, such as tension pneumomediastinum, delayed pneumothorax, airway compromise, and mediastinitis.

Conclusion: Spontaneous pneumomediastinum is a benign condition presenting primarily in young adults, with an uneventful recovery. Therefore, patients may recover from this condition without admission or the need for prophylactic antibiotics.

Background

Spontaneous pneumomediastinum (SPM) is defined as the presence of free air in the mediastinum without apparent concomitant factors or disease. It is uncommon but usually benign and self-limiting.

The pathophysiology of this disease is based on a pressure gradient between the alveolus and the lung interstitium. This pressure gradient may lead to alveolar rupture and the consequent flow of air into the in- terstitium. Once the air is in the lung interstitium, it flows towards the hilum and the mediastinum along a pressure gradient between the lung periphery and the mediastinum [1]. Representative computed to- mography (CT) images of SPM are shown in Fig. 1.

Generally, patients with SPM are admitted to hospital. Panacek et al. included only 3 of 17 patients (17.6%) evaluated on an outpatient basis [2]. In retrospective studies of SPM, some authors have reported the

* Corresponding author at: 2-1-1 Minatojima-Minamimachi, Chuo-ku, Kobe-city, Hyogo 650-0047, Japan.

E-mail address: [email protected] (M. Ebina).

mean length of hospital stay to be 2-8 days [3,4]. However, patients with SPM that is benign and self-limiting may not need to be admitted to hospital. Prophylactic antibiotics are occasionally administered in cases of SPM to prevent mediastinitis [5,6,7]. However, prophylactic an- tibiotic coverage in SPM is a controversial issue.

The purpose of this study was to describe practices concerning the feasibility of outpatient treatment and follow-up of SPM and the neces- sity of prophylactic antibiotics.

Materials and methods

We conducted this observational study in a single emergency med- ical center located in Kobe, Japan. This center includes a medico-surgical emergency department (ED) for adult and pediatric patients; 33 000 pa- tients visit the ED and 6600 patients are admitted per year. We retro- spectively evaluated patients with SPM from January 2007 to December 2014.

Diagnosis of SPM was based on the criteria of Macia et al., except for the patient’s age [4]. Patients had to fulfill the following inclusion

http://dx.doi.org/10.1016/j.ajem.2017.03.017

M. Ebina et al. / American Journal of Emergency Medicine 35 (2017) 1150–1153 1151

Fig. 1. Computed tomography (CT) scans showing spontaneous pneumomediastinum. Interstitial air surrounds the structures within the mediastinal pleura.

criteria: (1) the presence of a clinical picture consistent with pneumomediastinum and (2) pneumomediastinum diagnosed by confirmation of free air in the mediastinum on chest radiography or CT. Pediatric patients were also included. Patients were excluded if there was any evidence of a clear trigger for pneumomediastinum, such as perforation of the trachea or esophagus, iatrogenic factors (pneumomediastinum occurring after thoracic or cardiac surgery), chest wounds and injuries, infection by gas-producing germs, or any disease involving the neck or abdomen. The mode of selection of the pa- tients was the diagnosis in the emergency department. The authors, emergency physicians, reviewed the cases, and radiologists interpreted the CT scans. Major trauma was defined as an injury severity score (ISS) N 12 and minor trauma as an ISS <= 12 [8].

All case records were carefully reviewed considering the demo- graphic data, symptoms, precipitating events, diagnostic workup per- formed, use of prophylactic antibiotics, length of hospital stay, and complications. Complications included tension pneumomediastinum, delayed pneumothorax, airway compromise, and mediastinitis. The patients who were not admitted to hospital were followed up by tele- phone for at least 1 month.

The ethics board in our hospital approved the study design (IRB ap- proval number: zn170201).

Results

During the study period, 34 patients were included: 26 men (76.5%) and 8 women (23.5%), with a mean age of 19.7 years (range, 5 to 36 years). Twenty-nine patients (85.3%) underwent chest radiography and CT, while 5 patients (14.7%) underwent chest radiography only. One patient underwent bronchoscopy, and another patient underwent esophagography and esophagoscopy, but no additional findings, includ- ing esophageal and tracheobronchial ruptures, were detected.

The most frequently reported symptom was chest pain in 61.4% of the patients (21 of 34), followed by dyspnea in 29.4% (10 of 34) and pharyngalgia in 29.4% (10 of 34). In terms of the precipitating event, no specific trigger (developed at rest) was the most common finding in 44.1% of the patients (15 of 34). In addition, physical exercise, coughing (with or without upper respiratory infection), asthma attack, minor trauma, shouting, blowing one’s nose, and childbirth were the suspected precipitating events (Table 1).

Twenty-three patients (67.6%) were followed up on an outpatient basis, and 11 patients (32.4%) were admitted to the hospital, with a mean length of stay of 3.4 days. Principal reasons for hospitalization are listed in Table 2. The main reason for hospitalization was observa- tion of SPM in 6 patients, with a mean length of stay of 2.8 days. Five pa- tients were admitted for the treatment or observation of underlying disease (e.g., asthma attack).

Prophylactic antibiotics were orally administered to 2 patients. Both patients received ampicillin plus clavulanic acid for 3 days.

No patient developed complications, such as tension pneumomediastinum, delayed pneumothorax, airway compromise, and mediastinitis. Two patients were lost to follow-up by telephone, but none of the admitted patients had complications, except for 1 pa- tient who showed recurrence.

Discussion

In this series of 34 patients, 23 patients (67.6%) were followed up on an outpatient basis and prophylactic antibiotics were admin- istered to only 2 patients. None of the patients developed any complications.

SPM was initially described by Laennec in 1819, and was further char- acterized in a case series by Hamman in 1939 [7]. It is a Rare condition with an incidence of b 1:44,000 and often follows a benign course [4].

1152 M. Ebina et al. / American Journal of Emergency Medicine 35 (2017) 1150–1153

Table 1

Demographics and clinical presentation of patients with spontaneous pneumomediastinum.

Table 3

Summary of previous studies on spontaneous pneumomediastinum.

Parameter Value, n (%) or Mean (range)

Author n Outpatient

Hospital

Complications Prophylactic

Age (years)	19.7 (5-36)		treatment stay		antibiotics
Male gender	26 (76.5)	Our study	34 23	3.7	0 2 (5.9%)
Symptom		Macia [4]	41 0	5	0 –^c
Chest pain	21 (61.8)	Panacek [2]	17 3	2.5	0 –^c
Dyspnea	10 (29.4)	Campillo-Soto [14]	36 0	8.56	0 –^c
Pharyngalgia	10 (29.4)	Bakhos [10]	49 11	1.8	1 –^c
Neck pain	1 (2.9)	Perna [11]	47 0	3.5	1 –^b

Back pain 1 (2.9)

Hoarseness 1 (2.9)

Precipitating event		Abolnik [9]	25	0	6.3	0
Asthma attack	4 (11.8)	Koullias [6]	24	12	2	0

Exercise	4 (11.8)
Cough/URI	4 (11.8)
Minor trauma	3 (8.9)
Shouting	2 (5.9)
Nose blowing	1 (2.9)
Childbirth	1 (2.9)
None (at rest)	15 (44.1)
Diagnostic modality

CT 29 (83.5)

Bronchoscopy 1 (2.9)

Esophagram and esophagoscopy 1 (2.9)

Prophylactic antibiotics 2 (5.9)

Admission 11 (32.4)

Length of stay (days) 3.4 (2-6)

Complications 0

Mortality 0

Values are presented as mean (range) or number (percentage). URI, upper respiratory infection; CT, computed tomography.

SPM has been associated with trigger factors in approximately 75% of cases. The most common trigger factors are cough, physical exercise, and drug abuse [9]. In our patients, we found a weak association with trigger factors. However, the most frequent triggers were asthma attack, coughing due to upper respiratory infection, and physical exercise, sim- ilar to that reported previously.

There are several case series of SPM; however, only a few cases dem- onstrated poor outcomes. Takeda et al. suggest that 2 days may be a fea- sible duration for observational hospitalization [5]. On the other hand, Panacek et al., suggested outpatient treatment and follow-up in stable patients [2], although they included only a few patients who were followed up on an outpatient basis. Our study had a higher proportion of outpatient cases (Table 3), and the main reason for admission was the severity of underlying disease such as Severe asthma attack. Thus, our results support the feasibility of outpatient treatment and follow- up in stable patients, as suggested by previous studies [2,4,6].

Avoiding hospitalization of these patients leads to cost effectiveness. Taking into consideration the benign nature of this condition, only those cases where the diagnosis is in question or the underlying disease needs specific treatment should be considered for further diagnostic workup and admission.

Occasionally, there is a concern regarding differentiating SPM from pneumomediastinum secondary to esophageal perforation. Bakhos et al. suggest assuming a high index of suspicion in cases of pneumomediastinum secondary to esophageal perforation [10]. Age over 40 years; history of severe vomiting; abdominal tenderness on physical examination; elevated white blood cell count; and CT scan

Table 2

Principle reasons for hospitalization.

Reason Number (n)

Observation of spontaneous pneumomediastinum 6

Treatment of asthma attack 3

Treatment of respiratory failure due to influenza 1

Observation after childbirth 1

Total 11

Caceres [13] 28 –^a 3 0 –^c

Takada [5] 25 0 7.8 0 19 (76%)

–c

24 (100%)

Values are presented as number of patients or mean number of days of hospital stay.

^a The authors stated that the “majority of cases were admitted.”

^b The authors described the use of prophylactic antibiotics for the prevention of mediastinitis, but the number of patients who were administered antibiotics was not stated.

^c The authors did not discuss the use of prophylactic antibiotics.

findings of pleural effusion, significant atelectasis, pneumopericardium, or pneumoperitoneum are considered high risk factors; thus, when these findings are observed, patients with pneumomediastinum may require further diagnostic workup and admission.

Only a few small studies have reported complications in cases of SPM. Perna et al. reported a case series of 47 patients and 1 new case wherein surgical intervention was necessary [11]. In that case, a poor outcome at 3 h after admission, right Tension pneumothorax, mediasti- nal shift, and tracheal compression necessitated a right thoracotomy to open the mediastinal pleura. The clinical course was favorable; the chest tube was removed at 48 h and the patient was discharged on the third postoperative day. However, at the time of diagnosis, a pre-existing me- diastinal shift with moderate tracheal compression and right minimal pneumothorax were observed. Therefore, CT findings such as pneumo- thorax or mediastinal shift with moderate tracheal compression are in- dicators for admission.

Bakhos et al. reported a case series of 49 patients and only 1 patient requiring intubation for a severe asthma attack (pneumomediastinum was diagnosed before intubation). However, no mortality was recorded in the study population [10]. The treatment of underlying disease, as we report, was an exceptional reason for admission of patients with SPM.

Some previous studies recommend prophylactic antibiotics for preventing mediastinitis [5,6,12]. In the study by Koullias et al., all pa- tients were administered prophylactic antibiotics [6]. On the other hand, in our study, only 2 of the 34 patients were administered prophy- lactic antibiotics. Nevertheless, no patient developed infectious compli- cations including mediastinitis. In addition, we could not find evidence of patients with SPM who developed mediastinitis.

Thus, we can possibly avoid using prophylactic antibiotics in patients with SPM in terms of preventing the emergence of drug-resistant strains of bacteria. Moreover, this may contribute to cost saving and avoiding hospitalization.

Our study included 3 minor trauma patients. Generally, pneumomediastinum with major trauma such as penetrating injury is considered secondary pneumomediastinum. However, minor trauma such as falling on one’s hips can trigger a strong Valsalva maneuver, cre- ating high intrathoracic pressure. Therefore, pneumomediastinum with minor trauma has the same mechanism as SPM.

Limitations

Our study has some limitations. First, this was a retrospective obser- vational study, so we could not prove cause and effect, and our study might have lacked the power to detect complications. Second, 2 patients were lost to follow-up; thus, assessment of complications may be insuf- ficient, although the 2 patients accounted for only a small percentage of all patients. Third, the criteria for hospitalization depended on the

M. Ebina et al. / American Journal of Emergency Medicine 35 (2017) 1150–1153 1153

attending physician. Fourth, the reason for using prophylactic antibi- otics was unclear; nevertheless, only 2 patients were administered pro- phylactic antibiotics.

Conclusion

SPM is a benign condition presenting primarily in young adults, with an uneventful recovery. More than half of the patients were followed up on an outpatient basis, and prophylactic antibiotics were administered to only 2 patients. None of the patients developed any complications. Therefore, the study results support the feasibility of outpatient treat- ment and follow-up of SPM and the unnecessity of prophylactic antibi- otics. These patients should be considered for further diagnostic workup and admission only when the diagnosis is in question, the underlying disease needs specific treatment, or the possibility of organ perforation cannot be ruled out. However, our study might have lacked the power to detect complications, so further investigation is needed for clarifica- tion of these topics.

Funding sources

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

Conflict of interest

None.

References

AJEM