| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Click here to read this article as a CME activity
a Department of Surgery, Boonshoft School of Medicine, Wright State University, Dayton, Ohio
b Veterans Administration Medical Center, Dayton, Ohio
Accepted for publication January 12, 2009.
* Address correspondence to Dr Paull, Department of Veterans Affairs, National Center for Patient Safety, 24 Frank Lloyd Wright Dr, Ste M2100, Ann Arbor, MI 48106-0486 (Email: douglas.paull{at}va.gov).
Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5–8, 2008.
| GENERAL THORACIC SURGERY:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Data were collected from 104 consecutive patients with empyema. Primary outcomes were additional procedures and death. Predictor variables included age, delay, Karnofsky performance status (KPS), Charlson comorbidity index (CCI), serum albumin, malignancy, Acute Physiology and Chronic Health Evaluation II score, loculations on computed tomography scan, empyema stage, and first procedure choice.
Results: Advanced empyema (
stage IIA) was present in 84% of patients. Overall treatment success rates (no death, no additional drainage procedures) among evaluable patients for pigtail drainage, tube thoracostomy, VATS, and thoracotomy were 40% (4 of 10), 38% (14 of 37), 81% (13 of 16), and 89% (32 of 36), respectively. Five patients underwent miscellaneous procedures. Univariate variables associated with hospital death included KPS, CCI, and drainage as the first procedure. In multivariate analyses, KPS (coefficient, –0.06, p = 0.002) and failure of the first procedure (odds ratio [OR], 6.76; 95% confidence interval [CI], 1.45 to 31.4, p = .01) were independent predictors of death. Simple drainage as the first procedure was a strong, independent predictor of failure of the first procedure (OR, 11.1; 95% CI, 3.51 to 34.9; p = .00004).
Conclusions: The choice of the first procedure is critical in the outcome for treatment of empyema, even with adjustment for confounding variables. VATS or thoracotomy as initial therapy for advanced empyema is associated with better outcomes.
Thoracic empyema is a historical challenge for thoracic surgeons. The first recorded discussion of empyema and its treatment is found in the medical texts composed by Hippocrates and his school between the fifth and the second centuries BC [1]. These writings recommend and describe open drainage techniques for use when patients failed to respond to nonsurgical therapies. Interestingly, the affected chest was identified by shaking the patient and listening for a splashing or rippling sound [1]. Since these initial reports, the optimal treatment of patients with infected pleural effusions has been debated. Pleural infections develop in about 65,000 patients per year in the United States and the United Kingdom, and about 15% of these patients die and 15% to 40% undergo surgical drainage of the infected pleural space [2, 3].
Today, alternative strategies for the initial therapy range from antibiotic treatment (with or without thoracentesis) to drainage with radiologically placed pigtail catheters or tube thoracostomy to surgical intervention that can be performed using minimally invasive techniques, including video-assisted thoracic surgery (VATS) or open thoracotomy. In the hopes of contributing to the ongoing search for the best treatment strategy, we reviewed our recent experience at Wright State University Boonshoft School of Medicine Department of Surgery at 2 hospitals, the Department of Veterans Affairs (VA) Medical Center in Dayton and Miami Valley Hospital.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
To form a truly consecutive, nonselective, patient population, microbiology laboratory records were the primary means of identifying patients at both facilities. Any patient in whom a sample of pleural fluid was sent for culture triggered a review of the clinical record for possible inclusion in the study. We examined the study patient selection process. At the VA Medical Center, pleural fluid microbiologic specimens were submitted for 544 patients, and 50 (9%) with empyema were identified. This result correlated with the reported empyema prevalence of 5% to 20% for pleural effusions [4]. Information extracted from the records included demographic profile; clinical status; findings from radiologic studies, including chest roentgenogram and chest computerized tomography (CT) scans; results of pleural fluid analysis; time interval to treatment; specific empyema treatment(s); and outcomes, including patient survival, empyema resolution, complications, and length of stay.
We adapted the classic American Thoracic Society (ATS) classification system to better characterize ATS stage II patients, who represent a wide spectrum of disease and varied treatment approaches [5]. ATS stage III patients whose pleural space is encased by a thick, fibrotic cortex, do not constitute a therapeutic dilemma because they are not amenable to drainage treatment and require a true surgical decortication [6]. True fibrothorax, stage III patients were not the focus of our study. Patients were retrospectively staged by 2 board-certified thoracic surgeons (A. G. L. and D. E. P.) using pleural fluid biochemical data, radiographic studies, and operative findings [5–7].
We assigned patients to a modified ATS classification system as follows: Stage I was unchanged and defined an early stage of empyema with thin, free flowing fluid, and a pH exceeding 7.3. Stage II was subdivided into IIA and IIB, with the primary distinction being the presence or absence of a pleural peel. Stage IIA represented a fibrinopurulent empyema with loculations, minimal or no pleural peel, and a pH of less than 7.1. Stage IIB represented fibrinopurulent empyema with loculations, a significant pleural peel, and a pH of less than 7.1.
Operative steps, both VATS and open, included evacuation of all infected fluid, disruption of all loculations, débridement of pleural peels, and placement of 1 to 3 chest tubes. The goals were complete evacuation of infected material and restoration of complete lung expansion to eliminate any potential pleural space.
Data were entered into an Excel spreadsheet (Microsoft, Bellevue, WA). Data were analyzed using NCSS/PASS 2000 statistical software (NCSS Statistical Software, Kaysville, Utah) to determine the relationships between patient characteristics, stage of empyema, choice of first treatment, procedure success, and death. "Procedure success" was defined as the resolution of empyema with radiographic lung reexpansion, the absence of ongoing sepsis, and no additional drainage procedures. "Overall all treatment success" was defined as no death and no additional drainage procedures. Univariate data comparisons included the t test for continuous variables and the Fisher exact test for discrete variables. Data are reported as a mean ± standard error of the mean. A value of p < 0.05 was considered significant. The p values were corrected for multiple comparisons using the Bonferroni method. Multivariate analysis included logistic regression with death and procedure failure as dependent variables.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
|
|
|
|
Treatment strategies and outcomes were influenced by the clinical stage of empyema. Simple drainage was more frequently used for early-stage empyema, whereas surgical options were usually done for more advanced empyema (Fig 3A). Simple drainage as an initial treatment strategy became less effective for empyema at an advanced clinical stage (Fig 3B).
|
Univariate data analyses are summarized in Table 3. Significant predictors of "overall treatment success," defined as no death and no further drainage procedures, included Karnofsky performance status (KPS), Charlson comorbidity index, and choice of the first procedure (simple drainage vs operation). In a multivariate logistic regression, independent predictors of death included KPS and failure of the first procedure attempted. The strongest predictor of failure of the first procedure was choice of simple drainage. Overall, the single, best predictor of either death or the need for another procedure was simple drainage chosen as the first procedure (Table 4).
|
|
KPS was a particularly strong predictor of overall treatment failure and death (Tables 3 and 4). We hypothesized that the aspect of KPS that contributed to this finding may include its summary relationship to the presence of malignancy, other medical problems (Charlson comorbidity index), and the severity of illness (Acute Physiology and Chronic Health Evaluation [APACHE] II score) that a particular patient presented. We separated empyema patients into two groups, KPS of 60 or less and KPS exceeding 60. Patients with lower KPS were more likely to have an underlying malignancy (32% [15 of 47] vs 5% [3 of 57], p = 0.0003), higher Charlson comorbidity indices (3.9 ± 0.5 vs 1.7 ± 0.2, p = 0.00001), and elevated APACHE II scores (12.4 ± 0.9 vs 7.3 ± 0.4, p = 0.000005). KPS was a strong predictor of outcomes, but multivariate analysis revealed that choice of first procedure remained the best predictor of treatment failure (Table 4).
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Some have recommended surgical treatment only after simple drainage, with or without fibrinolytics, has failed [13, 14]. The current results emphasize the critical importance of selecting the proper procedure initially. Failure of the first procedure was the strongest predictor of hospital death, even when controlling for confounding host factors such as performance status and comorbidities. Although tube thoracostomy may successfully drain free flowing pleural fluid in the absence of loculations or peel, it is not sufficient in the treatment of more advanced empyema. VATS or thoracotomy allow proper débridement of pleural peels, disruption of all loculations, and evacuation of all fluid. These steps are essential to achieve full lung expansion that obliterates the pleural space and leads to resolution of the infectious process.
KPS was also a strong predictor of overall treatment failure or death. Our data demonstrated that patients with lower KPS were more likely to have an underlying malignancy and had more preexisting medical problems. Nonetheless, choice of first procedure to treat the empyema was an even stronger, independent predictor of overall treatment failure or death. Age was not an independent predictor of overall treatment failure or death. The relationship between younger age and worse outcomes with simple drainage as a first procedure for the treatment of empyema was explained by a greater proportion of trauma and postoperative patients included in the younger group.
Simple drainage was definitive treatment in more than 60% patients with stage I empyema. Surgical drainage was effective in all instances of stage I disease, but its routine use would expose a significant number of patients to the risks of anesthesia and an operation that might not have been necessary. Our study did not specifically set out to address which tools and techniques are best adapted in the initial staging of patients with empyema. However, lower pleural fluid pH and chest roentgenogram or CT findings of loculations or peel were associated with higher stage of empyema determined retrospectively using intraoperative findings. Other studies have suggested that CT scan and ultrasound imaging are not always reliable in the preoperative staging of empyema [6, 15]. Most patients in this study had ATS stage II empyema. Operation, whether VATS or thoracotomy, is the most successful initial procedure. These results are consistent with the conclusions of other studies that demonstrated greater success rates when VATS was the first-line therapy in the management of complex empyema [16, 17].
The retrospective nature of this study precludes knowing an individual surgeon's thought process when a given strategy was chosen, but it is likely that critical information considered included pleural fluid analysis, radiographic studies, and the patient's clinical course. The success of VATS has also diminished the role of fibrinolytics, which were used in a limited number of our patients. Although some authors have advocated fibrinolytics in selected parapneumonic empyema patients, others have reported poor results [18, 19]. Given the overall success of operative intervention both in our series and in previous reports, the equivocal utility of fibrinolytics, and the consequences of failed drainage, we do not advocate a significant role for fibrinolytics for patients with advanced empyema.
Thoracotomy was more commonly used than VATS in this study. This may have been due to inclusion of postoperative and traumatic causes of empyema where open procedures were more often required. Cause-specific death was higher among the patients with secondary or postoperative empyema. Also influencing the initial choice of thoracotomy was that most patients had advanced clinical stage empyema with evidence of an extensive peel upon presentation to the thoracic service.
We did not perform a cost-effectiveness analysis because we were focusing only on patient clinical outcomes. Applying VATS or thoracotomy for an early empyema that might resolve with tube thoracostomy and antibiotics could be more costly and risky. Other studies have demonstrated improved cost-effectiveness with early surgical intervention [20]. Although we did not determine hospital costs, prolonged hospital stays and significantly higher complications characterized patients in whom the initial drainage procedures failed, further underscoring the importance of choosing the correct initial procedure.
The current results are principally useful in indicating that the preferable initial treatment of patients with fibrinopurulent empyema, with (stage IIB) or without (stage IIA) an inflammatory peel, is surgical intervention rather than simple drainage procedures. This surgical intervention is most accurately described as drainage and débridement rather than decortication. Decortication is a procedure for an end-stage process with fibrous entrapment of the lung, described as a cortex (from the Latin meaning tree bark). Whether a thoracotomy or VATS is used for the surgical procedure does not seem to matter as long as all loculations are evacuated and inflammatory peels are removed to allow full lung expansion and obliteration of the pleural space. The futility of draining fibrinopurulent empyema with pigtail catheters is apparent in this and other studies [21].
In summary, the current study demonstrates that the choice of the first procedure to treat empyema has important implications for patient outcomes. Operative intervention with VATS or thoracotomy is the optimal initial approach for patients with advanced empyema [7].
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
When I think about empyema and effusion, I kind of think of it in two stages, and the first stage is whether that collection of fluid—that parapneumonic collection—needs to be drained in the first place or will it just go away on its own, and if we determine that it does need to be drained, what is the procedure that you do to drain it? Now, I think the way you get the opportunity to make these decisions or go through this thought process is if somebody involves you in the case earlier on, and if you are involved in the case later on, you oftentimes don't get to make that decision because it has already moved on to the later stages of empyema. And one of the things that I noted in your manuscript was there is a fairly long preprocedure length of stay.
My first question for you is, did you notice any difference in that preprocedure length of stay compared to the morbidity and the mortality once you did the procedure?
My second question is more of a comment. My interpretation of the literature about fibrinolytics is different than yours. You, I take from you paper, don't think they work. I think there is a place where they can work, but they need to be sort of carefully selected. Now, I may not actually get to make a choice, because last week I was told that we are out of urokinase in the United States. So, I just have to take them to operation anyway, but I wondered if you could maybe comment on why you think fibrinolytics have no purpose in the treatment of empyema. Thank you.
DR WOZNIAK: Thank you, Dr Schipper. I appreciate you looking at our paper and for your insightful comments. Your first question related to delay in referral to thoracic surgery. Prior to my involvement in the current study, my bias was similar to yours, that delay would be a predictor of outcome. In univariate analysis of our data comparing treatment successes with treatment failures, the delays were similar. Other studies have found delay to be associated with worse outcomes.
The second question related to the use of fibrinolytics. As you noted, there were only a few patients in the current study where fibrinolytics were utilized. In other studies, fibrinolytics have had mixed success in the treatment of empyema, and based on this it's our bias that fibrinolytics have a limited role. We agree with you that if fibrinolytics are considered, that decision should include input from the thoracic surgeon. This will help ensure proper patient selection as well as timely surgical intervention with VATS or thoracotomy for fibrinolytic failure.
DR CAROLYN E. REED (Charleston, SC): Let me just take the prerogative of the moderator here to ask you a question. At our institution it is usually a Friday, close to 5 o'clock in the afternoon, when you get an emergency call from the pulmonary department that they have this patient that has had at least two, maybe more, pigtails put in that haven't resolved the problem, and they are asking you to emergently take the patient to surgery to solve the problem. So my first comment is, I think it is rare that we actually get to make the choice; and second, I would say, on the strength of your data, we really need to get this word out to our pulmonary colleagues. Maybe you could comment on whether your university or hospital is better than mine in that regard.
DR WOZNIAK: I wish it were.
DR REED: You have answered the question.
DR STEPHEN CASSIVI (Rochester, MN): It was a very good presentation. After looking at your abstract and then listening to your presentation, I agree with you. The first go at it is always the best chance to deal with an empyema. But as we have just discussed, it is often difficult to get that first go at it, and we often get it much later on in the progression of the condition. For this disease I think it is very hard to figure out preoperatively who is a so-called IIA and who is a IIB. Although I agree there is probably not a lot of added cost with initially trying the VATS [video-assisted thoracic surgery] approach for the possible or presumed IIA case, I think once you get to the IIB stage I would argue that the VATS approach is not adequate. In terms of VATS versus open approaches, can you really do an adequate decortication for stage IIB empyema by VATS? As an aggressive minimally invasive surgeon, I don't want to sound like a Luddite, but I do appreciate the difficulty of decortication of late-stage empyemas by the VATS approach. The residents at our institution are taught to think of two realities when they are told of a patient receiving VATS decortication: one, the patient didn't need a true decortication; or, two, the patient didn't get a true decortication. I would be interested to hear your comments on these matters.
DR WOZNIAK: I agree that distinguishing preoperatively between a IIA and IIB may prove difficult. A CT [computed tomography] scan demonstrating a peel or loculations goes a long way in separating stage I from stage II, but not necessarily separating IIA from IIB. Pleural fluid analysis, especially pH, may also be important supplementary data to consider. Based on our data, one could argue that VATS for suspected stage II empyema is valuable. During VATS, one can evaluate the true stage of empyema. If the ultimate goal of evacuation of loculations, removal of all the purulent material, and lung re-expansion can be achieved, VATS is a good option. Otherwise, for more advanced cases, conversion to thoracotomy will be required. The large proportion of thoracotomies in our study was due, in part, to the large number of patients with advanced stage empyema.
In terms of decortication, my mentor, Dr Little, helped ensure we maintain a strict definition. Decortication via thoracotomy was carried out for a stage III empyema, dense adhesions with fibrothorax. The current study excluded stage III patients. We agree with your characterization of the current role for VATS and decortication for such patients.
DR JOHN A. HOWINGTON (Evanston, IL): Actually just a comment. We did a prospective analysis at the University of Cincinnati that has been submitted—and hopefully will be published this year—using fibrinolysis but working with our interventional colleagues. We got to patients immediately. The patients had a chest tube placed. If the effusion wasn't fully drained, it was a IIA. They got fibrinolytics for 48 hours. If it wasn't better, they got operated on. No deaths. Eighty percent were taken care of with fibrinolytics. All others could be managed with thoracoscopy. So it can happen.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
