Utility and Safety of Decortication Using Medical Thoracoscopy under Local Anesthesia Performed by Respiratory Physicians in Patients with Acute Pleural Empyema: A Single-center Retrospective Study

Abstract

Objective

Acute pleural empyema, defined as empyema diagnosed within three months of the onset, frequently affects patients with poor general health and often poses challenges for invasive treatments requiring general anesthesia. In addition, the number of respiratory surgeons was insufficient and unevenly distributed across urban base and core hospitals in Japan. Therefore, safe and effective therapeutic procedures that physicians can perform are required.

Methods

We retrospectively analyzed 28 patients with acute pleural empyema who underwent decortication using medical thoracoscopy under local anesthesia (DMT-LA) in our department. All procedures were performed by respiratory physicians at our institution. The “success” of DMT-LA was defined by meeting all of the following criteria: (i) absence of residual clinically significant pleural effusion, (ii) resolution of clinical signs of sepsis, (iii) normalization of serum C-reactive protein (CRP) levels and white blood cell counts, (iv) no requirement for additional interventions, and (v) maintenance of these conditions for at least one month following antibiotic treatment.

Results

Of the 28 patients, 21 met the criteria for success, yielding a success rate of 75.0%. Patients in the failure group were more likely to have an Eastern Cooperative Oncology Group-performance status of ＞3, stage III empyema, and high median serum CRP levels than those in the success group. Adverse events were observed in only one patient (3.4%) and involved prolonged sedative effects.

Conclusion

DMT-LA is a safe and effective therapeutic procedure that can be performed by respiratory physicians and provides a viable treatment option for acute pleural empyema, particularly in facilities without respiratory surgeons.

Introduction

Acute pleural empyema is characterized by the accumulation of pus in the pleural cavity within three months of the onset. Antibiotic administration and drainage of pus are the primary treatments for acute pleural empyema; however, surgical interventions under general anesthesia, such as video-assisted thoracoscopic surgery or open thoracotomy, are recommended when the condition does not respond to initial treatments (1-3). Nevertheless, acute pleural empyema commonly affects older adult patients or those with comorbidities (4,5), making invasive treatments challenging due to the risks associated with general anesthesia. Therefore, there is a need for safer therapeutic procedures, particularly in high-risk patients.

Furthermore, the distribution of respiratory surgeons in Japan is uneven, with a notable shortage in non-urban areas and hospitals with limited medical resources. Consequently, many hospitals lack respiratory surgeons, especially in resource-limited regions. Respiratory physicians play a critical role in managing acute pleural empyema. Therefore, therapeutic approaches that can be safely used by respiratory physicians are required.

Our institution, Municipal Tsuruga Hospital, is situated in Fukui Prefecture, one of the least populated prefectures in Japan. Furthermore, Tsuruga City, where our institution is located, has limited medical resources. As a regional center for emergency medical care, we receive numerous emergency patients and ambulances, including patients with respiratory diseases such as acute pleural empyema. However, as there are no respiratory surgeons at our institution or nearby facilities, patients requiring thoracic surgical interventions must be transported to higher-level medical centers for over an hour.

In light of this current situation, we, as respiratory physicians, have performed decortication using medical thoracoscopy under local anesthesia (hereinafter referred to as “DMT-LA”) for acute pleural empyema on admission or the following day in all cases since April 2020, aiming to maximize cure rates at our institution. Based on our experience, we believe that the DMT-LA is both safe and useful. However, few studies have examined the utility and safety of this procedure when performed by respiratory physicians instead of respiratory surgeons. Therefore, this retrospective study evaluated the utility and safety of DMT-LA performed by respiratory physicians.

Materials and Methods

1．Study design and patients

This retrospective study was conducted at the Municipal Tsuruga Hospital and approved by the Ethics Committee of the Municipal Tsuruga Hospital (Approval Number 957, approved on March 13, 2024). This study complied with the principles of the Declaration of Helsinki, and patients were allowed to opt out.

We retrospectively analyzed 66 patients who underwent medical thoracoscopy under local anesthesia at our institution between April 2020 and March 2024. Among these, 27 were diagnosed with malignant pleuritis and 11 with pleuritis unrelated to empyema (including one parapneumonic pleural effusion). Therefore, these patients were excluded from the analysis. We focused on the remaining 28 patients who underwent DMT-LA for acute pleural empyema and collected clinical data from their medical records. Acute pleural empyema was defined as empyema diagnosed within three months of the onset of clinical signs of infection, such as a fever or cough. Furthermore, negative culture and polymerase chain reaction results for Mycobacterium tuberculosis were confirmed in all cases.

2．Procedure of DMT-LA

The DMT-LA procedure was performed as follows: sufficient intrapleural space was confirmed using ultrasound, and a marker was placed after verification. Lidocaine (1%) was injected locally for analgesia, and midazolam and fentanyl were administered intravenously at the discretion of the attending physician for local anesthesia before a skin incision. After skin incision with a scalpel and dissection of the intercostal muscles with pean forceps, a flexible thoracic trocar with an 8 mm diameter (MAJ-1058; Olympus, Tokyo, Japan) was inserted into the pleural cavity. In all patients, the procedure was performed using a single port.

The pus that accumulated in the pleural cavity was aspirated using a flex-rigid pleuroscope (LFT-H290; Olympus), and the fibrin net was debrided as much as possible using disposable biopsy forceps (FB-231D; Olympus) (Fig. 1). All procedures were performed by respiratory physicians at our institution before insertion of the trocar catheter into the pleural cavity.

Figure 1.

Two representative cases of patients who underwent decortication using medical thoracoscopy under local anesthesia (DMT-LA). The upper and lower panels represent cases of success and failure, respectively. Computed tomography (CT) findings are shown in the left panels. Thoracoscopic findings at the start of the procedure (second panels), during the procedure (third panels), and at the end of the procedure (right panels) are shown.

After the procedure, a trocar catheter was placed in the pleural cavity, which was rinsed daily with a saline solution. The timing of trocar catheter removal and duration of antibiotic administration were determined by each attending physician based on radiological and hematologic findings.

3．Outcomes

Based on a previous study (6), we defined the “success” for DMT-LA when all of the following criteria were met in this study: (i) absence of residual clinically significant pleural effusion as determined by a radiological assessment; (ii) resolution of clinical signs of sepsis, including fever, tachycardia, and/or hypotension; (iii) normalization of serum C-reactive protein (CRP) levels and white blood cell counts to baseline values; (iv) no requirement for additional interventions, including surgical treatments; and (v) maintenance of these conditions for at least one month after completion of antibiotic treatment. Conversely, “failure” was defined as failure to meet any of the above criteria. Patient characteristics and procedural details, including procedure time, were compared between the success and failure groups.

The procedure time for DMT-LA was defined as the time from skin incision to removal of the pleuroscope following debridement.

4．Statistical analysis

Differences in clinical features between the success and failure groups were statistically analyzed using Fisher's exact test and Mann-Whitney U test. The best cut off values were determined using a receiver operating characteristic (ROC) curve analysis. Statistical significance was set p values of ＜0.05. Statistical analyses were performed using the EZR statistical software program, version 1.55 (Jichi Medical University Saitama Medical Center, Saitama, Japan).

Results

1．Patient characteristics and success rate

Patient characteristics are shown in Table 1. Of the 28 patients, 21 met the definition of success. Among the seven patients in the failure group, the main reasons for failure were the need for surgical treatment in addition to DMT-LA because clinical sings of sepsis could not be resolved in five patients and the death of two patients due to pleural empyema.

Table 1.

Patient Characteristics.

	All n=28	Success n=21	Failure n=7	p value
Median age, years (range)	72 (47-87)	71 (47-87)	74 (53-84)	0.48
Sex
Male	21	15	6	0.64
Female	7	6	1
Median BMI, kg/m2 (range)	22.1 (14.5-29.5)	22.3 (14.5-29.5)	20.7 (18.4-23.0)	0.23
ECOG-PS
0-2	21	19	2	0.003
3-4	7	2	5
Stage of empyema
II	23	20	3	0.008
III	5	1	4
Respiratory failure
Yes	11	8	3	1.00
No	17	13	4
Etiology
Postoperative pleural empyema	0	0	0	-
Comorbidities
Diabetes mellitus	6	3	3	0.14
Cerebrovascular disease	6	3	3	0.14
Chronic obstructive pulmonary disease	4	3	2	0.57
Cardiovascular disease	5	2	2	0.25
Malignancy	2	2	0	1.00
Oral disease
Yes
Periodontitis	18	13	5	-
Tooth cavities	10	8	2
No	5	4	1
Unknown	5	4	1
Blood test
Median peak WBC, /µL (range)	20,200 (9,300-40,200)	19,800 (9,300-40,200)	22,800 (11,200-35,100)	0.94
Median peak CRP, mg/dL (range)	24.8 (15.0-61.1)	22.6 (15.0-34.6)	33.3 (21.5-61.1)	0.008
Median nadir albumin, g/dL (range)	2.4 (1.5-3.4)	2.5 (1.6-3.4)	2.3 (1.5-2.5)	0.06
Pleural fluid analysis
Median pH (range)	7.20 (6.3-8)	7.30 (6.3-7.8)	7.25 (6.3-7.6)	0.74
Median total cell counts, /µL (range)	4,366 (81-312,220)	5,191 (81-248,732)	4,161 (2,448-4,729)	0.50
Median LDH, U/L (range)	1,132 (21-20,000)	1,119 (391-20,000)	1,791 (21-4,475)	0.93
Median glucose, mg/dL (range)	54 (1-208)	56 (1-208)	27.5 (1-98)	0.49
Pleural fluid bacteriology
Negative	13	8	5	0.20
Positive	15	13	2
Streptococcus intermedius	10	10	0
Streptococcus constellatus	2	1	1
Streptococcus anginosus	1	1	0
Fusobacterium nucleatum	1	1	0
Pseudomonas aeruginosa	1	0	1

BMI: body mass index, ECOG: Eastern Cooperative Oncology Group, PS: performance status, WBC: white blood cell, CRP: C-reactive protein, pH: potential hydrogen, LDH: lactate dehydrogenase

Key differences between the success and failure groups included the proportion of patients with Eastern Cooperative Oncology Group (ECOG)-performance status (PS) of ＞3 being significantly higher in the failure group [71.4% (95% confidence interval (CI): 29.0-96.3%)] than in the success group [9.5% (95% CI: 1.2-30.4%)] (p=0.003), and the proportion of stage III according to the American Thoracic Surgery classification (7) being significantly higher in the failure group [85.7% (95% CI: 42.1-99.6%)] than in the success group [4.8% (95% CI: 0.1-23.8%)] (p＜0.001). In addition, the median peak serum CRP levels before the procedures were significantly higher in the failure group [33.3 (range: 21.5-61.1) mg/dL] than in the success group [22.6 (range: 15.0-34.6) mg/dL] (p=0.008). Furthermore, the median nadir serum albumin level tended to be lower in the failure group [2.3 (range: 1.5-2.5) mg/dL] than in the success group [2.5 (range: 1.6-3.4) mg/dL]; however, this difference was not statistically significant (p=0.06).

The overall success rate was 75.0% (95% CI: 55.1-89.3%) (Fig. 2A, left bar). When limited to patients with stage II empyema, the success rate was 87.0% (95% CI: 66.4-97.2%). In contrast, the success rate in patients with stage III empyema was 20.0% (95% CI: 0.5-71.6%) (p=0.008) (Fig. 2A, middle and right bar).

Figure 2.

(A) Association between success rate and stage of empyema. (B) Association between the success rate and peak serum C-reactive protein (CRP) levels using a cutoff value of 32 mg/dL. (C) Association between the success rate and a composite parameter comprising the stage of empyema and peak serum CRP levels using a cutoff value of 32 mg/dL.

2．Cutoff values for serum CRP levels as predictive markers of success

Given these differences, we focused on peak serum CRP levels as a predictive marker. An analysis using the ROC curve revealed that a peak serum CRP level of 32 mg/dL was the optimal cutoff value for predicting success (Fig. 3). Using this cutoff value, the success rate was significantly higher in patients with a peak serum CRP level of ≤32 mg/dL [18 out of 20 patients; 90.0% (95% CI: 68.3-98.8%)] than in those with levels ＞32 mg/dL [3 out of 8 patients; 37.5% (95% CI: 8.5-75.5%)] (p=0.009) (Fig. 2B).

Figure 3.

Receiver operating characteristic (ROC) curve for predicting success based on peak CRP levels.

Next, we examined the association between success and a composite parameter of empyema stage and peak serum CRP level with a cutoff value of 32 mg/dL. As shown in Fig. 2C, the success rates were as follows: 94.4% (95% CI: 72.7-99.9%) for the stage II and CRP ≤32 mg/dL group, 60.0% (95% CI: 14.7-94.7%) for the stage II and CRP ＞32 mg/dL group, 50.0% (95% CI: 1.3-98.7%) for the stage III and CRP ≤32 mg/dL group, and 0% (95% CI: 0-70.8%) for the stage III and CRP ＞32 mg/dL group. The difference between the stage II and CRP ≤32 mg/dL and stage III and CRP ＞32 mg/dL groups was significant (p=0.003).

3．Details of procedure, safety, and post-procedural course

Procedural details are provided in Table 2. The median doses of midazolam and fentanyl administered were 2.0 (range: 1-5) mg and 20.0 (range: 10-71) μg, respectively. The median procedure time was 58 (range: 23-88) min. No significant differences in these parameters were observed between the success and failure groups. Adverse events related to DMT-LA occurred in only 1 patient [3.4% (95% CI: 0.1-17.8%)], consisting of prolonged sedative effects.

Table 2.

Details of the Procedure of Decortication Using Medical Thoracoscopy under Local Anesthesia (DMT-LA).

	All n=28	Success n=21	Failure n=7	p value
Median total dose of midazolam, mg (range)	2.0 (1.0-5.0)	2.0 (1.25-5.0)	1.5 (1.0-2.0)	0.15
Median total dose of fentanyl, μg (range)	20.0 (10.0-71.0)	22.5 (12.5-50.0)	15.0 (10.0-71.0)	0.47
Procedures time, min	59 (26-88)	60 (26-80)	48 (27-88)	0.83
Adverse events associated with the procedure
Pneumothorax	0 (0)	0	0
Bleeding requiring intervention	0 (0)	0	0
Other
Prolonged effects of sedative drugs	1 (4)	0	1

In the success group, the median duration of trocar catheter insertion after DMT-LA was 10.5 (range: 4-21) days, and the median duration of antibiotic administration after DMT-LA was 35.5 (24-105) days.

Discussion

In this study, we investigated the utility and safety of DMT-LAs performed by respiratory physicians in patients with acute pleural empyema. The strength of this study lies in its focus on DMT-LAs performed by respiratory physicians rather than respiratory surgeons. Our findings indicate that DMT-LA performed by respiratory physicians demonstrates a relatively high success rate and a low incidence of adverse events.

First, we demonstrated that DMT-LA achieved a relatively high success rate of 75.0% for acute pleural empyema even when performed by respiratory physicians. Given that the failure rate of primary treatments, including antibiotic administration and drainage, is approximately 33% (6), the success rate observed in this study supports its utility as a primary treatment option. Furthermore, Ohuchi et al. (8) reported a success rate of 79.3% for DMT-LA performed by respiratory surgeons under similar circumstances. This comparable success rate suggests that respiratory physicians can perform DMT-LAs with a quality similar to that achieved by respiratory surgeons.

We demonstrated that an ECOG-PS of 0-2, low serum CRP levels, and non-stage III empyema were predictive factors for success. Given that these factors worsen as pleural empyema progresses, our findings indicate that decortication should be performed as early as possible during the treatment process. For example, Hardavella et al. (9) reported a success rate of 84.5% for DMT-LA performed by respiratory surgeons for early-stage empyema (stage I or II). This success rate was higher than that reported in studies involving patients with stage III empyema, such as our study and a previous study by Ohuchi et al. (8), supporting our recommendation for early intervention. Owing to its high success rate, we strongly recommend DMT-LA, especially in patients with stage II empyema and peak serum CRP levels ≤32 mg/dL.

Second, we observed that DMT-LA is a highly safe treatment for pleural empyema, with an adverse event occurrence rate of only 3.4%, which is comparable to previous studies (8,9). The only adverse event observed was a prolonged sedative effect that was not directly attributable to DMT-LA. Based on our experience, DMT-LA can be performed safely if the decortication is limited to fibrin nets. Conversely, we refrained from decorticating lesions with significant adhesions or those prone to bleeding, as the safety of DMT-LAs cannot be guaranteed without the cooperation of respiratory surgeons. The success rate in patients with stage III empyema, particularly in those with peak serum CRP levels ＞32 mg/dL, was remarkably low. Therefore, immediate consultation with respiratory surgeons is recommended. Nevertheless, we recommend DMT-LA as the primary treatment before trocar catheter insertion, as the procedure can be performed using an incision of the same size required for trocar catheter insertion.

Additional advantages were noted to with the DMT-LA when performed by respiratory physicians. First, visualization of pleural empyema facilitates the explanation of the condition to patients and consults with respiratory surgeons. Furthermore, it provides training opportunities for young respiratory physicians to better understand the pathogenesis of pleural empyema. Second, it allows the confirmation of the optimal position for trocar catheter insertion. Third, it confirms the absence of other complications such as thoracic malignancies. Fourth, even in cases where pleural fluid cannot be adequately collected via thoracentesis because of the presence of numerous septations in the pleural cavity, thoracoscopy can effectively collect pleural fluid and may lead to the identification of pathogenic bacteria. Finally, as previously reported (10,11), thoracoscopic observation of the pleural cavity can lead to the identification of new pathological conditions. These advantages highlight the utility of medical thoracoscopy for examining or managing pleural effusion.

However, this study had certain limitations. First, this was a single-center retrospective study with a relatively small sample size. Second, we did not compare DMT-LA with primary treatment without DMT-LA. To evaluate the utility and safety of DMT-LA more objectively, large-scale prospective comparative studies with clearly defined protocols, including surgical timing, are required. Nevertheless, this study provides important insights into the potential of DMT-LA performed by respiratory physicians.

Conclusion

DMT-LA is an effective and safe therapeutic procedure that can be performed by respiratory physicians and should be considered as a standard treatment option for patients with pleural empyema. Respiratory physicians, particularly those working in facilities without respiratory surgeons, should incorporate DMT-LAs into their treatment strategies.

The authors state that they have no Conflict of Interest (COI).

Koki Nakashima and Takayuki Azuma contributed equally to this work.