Mini-thoracotomy and chest tube insertion for children with empyema

J THoRAc

CARDIOVASC SURG

84:497-504, 1982

Mini-thoracotomy and chest tube insertion for children with empyema In contrast to the tI\'(} prior decades, empyema in children has become a rare disease during the past 10 years. Thirty-two children with empyema from all causes were treated at The Children's Memorial Hospital in Chicago in the 11 years between 1970 and 1982. Twenty cases followed postinfectious pneumonia, most commonly due to Staphylococcus aureus. The remaining patients had either an esophageal leak or a pleural infection following an intrathoracic operation. Fourteen of our patients were managed with a "mini-thoracotomy," which allowed accurate drainage of purulent material, debridement offibrinous exudate. and rapid expansion of the lung with prompt relief from fever and toxicity. In our hands. this procedure provided results which were superior to those obtained with simple closed chest tube drainage. In addition, during this period of time, ultrasound examination of the chest has provided a rapid. accurate technique for the early detection and localization offluid accumulations within the pleural cavity. Pre-drainage thoracentesis can be accurately guided by ultrasonic imaging.

John G. Raffensperger, M.D., Susan R. Luck, M.D. (by invitation), Arnold Shkolnik, M.D. (by invitation), and Richard R. Ricketts, M.D. (by invitation), Chicago, Ill.

Empyema in children has become a rare disease since the epidemic of staphylococcal infections during the 1950s. However, empyema continues to complicate an occasional case of staphylococcal, hemophilus, or pneumococcal pneumonia. When the initial pneumonia is inadequately managed with antibiotics, a pleural effusion forms which, if unrecognized, becomes thickened with fibrinous material. Delayed diagnosis and inadequate drainage will prolong hospitalization and increase morbidity. Accurate interpretation of chest radiographs remains the key to diagnosis, but we have increasingly utilized ultrasound to define the nature of the pleural opacity and to localize persistent pockets. An early empyema may be treated with thoracentesis or the closed chest insertion of a tube. Either of these procedures is useless if the pus has become thick and loculated. Furthermore, the insertion of a chest tube into a crying, stubborn child can be frustrating for the surgeon and painful for the patient. A "mini-thoracotomy" with resection of a short segment of rib allows

complete evacuation of thick, fibrinous exudate and accurate placement of a large-bore tube. This procedure can be performed rapidly with the patient under general anesthesia. We have used this technique successfully in 14 children. It compares most favorably with our experience using closed tube thoracostomy.

Patients and methods

From the Departments of Surgery and Radiology, The Children's Memorial Hospital, and Northwestern University Medical School, Chicago, Ill. Read at the Sixty-second Annual Meeting of The American Association for Thoracic Surgery, Phoenix, Ariz., May 3-5, 1982. Address for reprints: John G. Raffensperger, M.D., 2300 Children's Plaza, Chicago, III. 60614.

Thirty-two patients required drainage of empyema at the Children's Memorial Hospital from July, 1970, through February, 1982. Some children with infected pleural fluid aspirated for diagnosis may have been omitted; these required no further treatment. Twenty cases followed acute infectious pneumonia, five were the sequelae of intrathoracic operations, and seven resulted from esophageal anastomotic leak or perforation (Table I). Adequate pleural drainage was accomplished by closed tube thoracostomy in seven, mini-thoracotomy and closed drainage in 14, formal thoracotomy with decortication in seven, and open drainage in four. Our experience with open pleural debridement and drainage has been confined to those patients with postpneumonic empyema. The charts of these 20 patients were reviewed with specific attention to indications for operative drainage and postoperative morbidity. Twelve patients (63%) had Staphylococcus aureus infections, and they formed a homogenous group for comparison. These

© 1982 The C. V. Mosby Co.

497

0022-5223/82/100497+08$00.80/0

The Jou mal of

498 Raffensperger et al.

Thoracic and Cardiovascular Surgery

Table I. Cause of empyema in 33 children Cause

Postpneumonic Staphylococcus aureus Hemophilus influenrae Streptococcal pneumoniae Escherichia coli Eikenella corrodens Anaerobic organisms Alpha Streptococcus Postoperative Staphylococcus aureus Streptococcus pneumonia Hemophilus influenzae and Eikenella corrodens Eosphageal perforation Enteric organisms

No. 12 3 I I I

I I

3 I

lavaged with saline or an antibiotic solution, and a large-bore chest tube is placed under direct vision in a suitable location. The tube is tunneled subcutaneously to a separate skin exit site. Complete drainage was quickly obtained and the lung rapidly expanded in all cases. The average patient was afebrile, with the rectal temperature consistently at or below 100.4 0 F or 380 C, and nontoxic by the third postoperative day. Two children had postoperative abdominal distention and prolonged ileus. One of these remained intubated for 36 hours after anesthesia. All other patients were extubated immediately.

I

Results

7

Adequate drainage was delayed beyond 4 to 5 days after presentation in nine patients because of failure to appreciate the significance of persistent opacity on chest radiographs. In four patients with persistent localized opacity, the fever and other symptoms subsided on specific antibiotic therapy only to recur when the antibiotics were discontinued. Adequate definitive drainage was delayed by 1 to 19 days in five other patients when initial thoracentesis or tube thoracostomy clearly failed to resolve the inflammatory picture. Prehospitalization oral antibiotics did not interfere with diagnostic evaluation because no patient had received an antibiotic effective against the infecting organism. Children whose staphylococcal empyema was drained through a mini-thoracotomy became afebrile and nontoxic more quickly than those treated by tube thoracostomy even though drainage was effected in the latter group earlier in the course of the disease (Table II). In eight cases the empyema was drained by thoracotomy after 6 to 38 days of illness; the patients were afebrile in 1 to 6 days (average of 2V2 days). Tube thoracostomies were performed in four patients between the third and ninth days; this group remained febrile for 8 to 11 days (an average of 10 days) after tube insertion. All children with nonstaphylococcal empyema required a thoracotomy (Table III). Tube thoracostomy had failed in two. The mini-thoracotomy was successful in six patients. Although three became afebrile very soon postoperatively, three others remained febrile for 2 weeks. The unusual clinical histories of the remaining two patients prompted different drainage procedures. In retrospect, equally good results probably would have followed a mini-thoracotomy. Underlying tumor was suspected in a 14-year-old white boy with an 8 month history of recurrent right lower lobe pneumonia and weight loss. When he presented with acute pleuritic pain and a large right pleural effusion, multiple biopsies and a partial decortication were performed through

children usually presented with a 3 to 5 day history of a febrile upper respiratory tract infection; six had abdominal pain; and two had febrile convulsions. Most were lethargic and had respiratory distress. Their ages ranged from 4 months to 3% years, with an average and median age of 16 months. Eight of the 12 were admitted or transferred to our hospital within the first week of illness. At this time, physical examination revealed signs of bronchopneumonia and chest x-ray films usually showed opacity compatible with a large loculated effusion, sometimes with multiple pockets of air. Thoracentesis was promptly attempted in nine children; diagnostic fluid was obtained in six. Children infected with other organisms tended to be older and to have additional medical complications. Hemophilus influenzae empyema was treated in three children; one had meningitis and another, purulent pericarditis. The only immunosuppressed patient in this series had Wiscott-Aldrich syndrome, which was managed with prednisone. He developed necrosis and perforation of the transverse colon complicated by pneumonia and an Escherichia coli empyema. Sixteen of the 20 patients with postpneumonic empyema were treated by operative drainage, 14 with a mini-thoracotomy (Fig. 1). The procedure is performed with the patient under general anesthesia. The area of suspected empyema can be precisely localized and marked by percussion, radiographs, or ultrasound. Needle aspiration may aid in localization of fluid. A 4 to 5 cm incision is made directly over an overlying rib. A 2 to 3 em section of rib is resected, and the empyema is entered through the periosteal bed. The incision is only large enought to insert one or two fingers and a suction tip in order to break up loculations and to remove thick, fibrinous material. The pleural cavity is

Volume 84

Mini-thoracotomy for empyema 499

Number 4 October, 1982

Fig. 1. Technique of mini-thoracotomy for empyema drainage. A, Child positioned. Empyema position marked with ultrasound guidance; needle aspiration for confirmation. B1, Short subperiosteal section of overlying rib resected. B2, Empyema entered through periosteal bed. C, Finger dissection breaks up loculations; fibrinous exudate removed with suction and irrigation. D, Large-bore chest catheter placed through periosteal incision, tunneled subcutaneously to separate exit wound.

a formal thoracotomy incision. Cultures of both the bronchial aspirate and pleural fluid were positive for Eikenella corrodens. Postoperative hemorrhage necessitated reexploration, but the boy became afebrile after 4 days. A 14-year-old black boy complained of rightsided chest pain and a productive cough 13 years following right middle lobectomy for bronchiectasis. An

alpha streptococcal empyema was managed by open drainage. Subsequent evaluation has shown segmental right lower lobe bronchiectasis. Discussion

During the nationwide epidemic of empyema due to phage type 80-81 Staphylococcus, many children were

The Journal of Thoracic and Cardiovascular Surgery

5 00 Raffensperger et al.

Table II. Postpneumonic staphylococcal empyema Duration of illness prior to drainage

Age 4mo 4mo 8 mo 10 mo 15 mo 16 mo 2 yr 2 yr 13 mo 17 mo 2'12 yr 3\12 yr

No. of toxic days following drainage

Patients treated with mini-thoracotomy I Tube thoracostomy I day preop. unsuccessful; multiple loculations 3 Bilateral empyemas drained 6 5 ? Febrile antibiotic reaction with erythema multiforme on postop day 3 I 3 I I

7 days 5 wk

9 days 15 days 7 days 6 days 18 days 29 days 4 3 4 9

days days days days

Complicating circumstances

Patients treated with closed tube thoracostomy 10 Empyema recurred after tube removed on day 4; second tube placed 8 10 II

Recurrent collection; chest tube repositioned and replaced twice

Table III. Postpneumonic empyema, with organisms other than Staphylococcus: Treated with mini-thoracotomy Age

Infecting organism

Duration of illness No. of toxic days prior to drainage following drainage

10 mo

Hemophilus infiuenzae

25 days

5 yr

Hemophilus influenzae

26 days

14

7 yr 7 yr

Hemophilus infiuenzae Streptococcus pneumoniae

5Y2 wk 14 days

I 2

6 yr

Escherichia coli

10 days

II

Bacteroides jragilis

14 days

14

II yr

effectively treated with antibiotics and closed tube thoracostomy. Some required conversion to open thoracostomy or a decortication. t , 2 Since then, more effective antibiotics have practically eliminated this disease. Physicians educated in the past decade may fail to recognize this complication of bacterial pneumonia. Ineffective antibiotic therapy administered on an outpatient basis or during the initial period of hospital admission allows progression of the disease. The child's symptoms are repressed and the purulent material organizes and becomes fibrinous. In our series of cases, ampicillin was most frequently prescribed as the initial antibiotic. All of the staphylococcal and hemophilus organisms cultured from these patients were ampicillin-resistant. With delay in effective therapy, the pleural cavity becomes obliterated by loculated pus, which is often too

Complicating circumstances Hemophilus infiuenzae meningitis; recurrent collection and fever 5 days after tube removed; mini-thoracotomy 22 days after chest tube initially placed Hemophilus infiuenzae purulent pericarditis; recurrent empyema drained 21 days after pericardial window and right thoracotomy Tube thoracostomy 7 days preop. unsuccessful; multiple loculations Wiscott-Aldrich syndrome; on steroids; presented with pneumonia and colon perforation; bowel resection and colostomy 7 days before pleural drainage Bronchopleural fistula

viscous for evacuation by thoracentesis or chest tube. Even when tube drainage of an empyema is eventually successful, there may be a prolonged period of fever and toxicity. Frequently, chest roentgenograms demonstrate ' 'thickened" pleura for weeks or months. 3 Multiple thoracenteses and the insertion of a tube into a child, even with sedation, local anesthesia, and great patience on the part of the surgeon, are difficult and painful procedures. Early diagnosis of empyema need not rest only on repeated chest roentgenograms or thoracenteses. Ultrasonography has been used to accurately delineate pleural effusions and collections. 4 The examination can be performed rapidly without sedation, and at the bedside if necessary. A pocket of exudate or pus can be precisely located and the site marked. This diagnostic mo-

Volume 84 Number 4

October, 1982

Mini-thoracotomy for empyema

50 I

5

Fig. 2. A, Preoperative radiograph of a 2-year-old boy with staphylococcal empyemaon the twenty-eighth day of illness. Pleural effusion was first seen on the seventh day. and he became afebrile during a 10 day course of intravenous moxalactam. Temperature spikes recurred I week after the boy started an oral cephalosporin. B, Ultrasound outlining anterior empyema cavity lateral to right nipple under the third and fourth ribs. C. Chest radiograph 4 weeks after drainage through the third rib.

dality can guide the surgeon to confident and early drainage . It is particularly useful in cases in which the chest radiograph or the clinical course is atypical (Fig. 2). Ultrasound examination of the chest was performed in five of our patients. The empyema cavity was clearly demon strated in all, in one patient as early as the ninth day of illness. A single cavity may be outlined or loculations of mixed fluid-solid elements recognized. Ultrasonography can provide accurate guidance for diagnostic thoracentesis and , at the same time, can minimize the risk of inadvertent needle puncture of the liver, spleen, diaphragm, or aerated lung. The incision employed for a mini-thoracotomy

drainage is no larger than that used for an open lung biopsy; however, the resection of 2 to 3 cm of a rib allows thorough debridement of thick, fibrinous material which would not drain through a tube. Although we insert a large chest tube through the periosteal incision , there is often little postoperative drainage and the lung expands immediately (Fig. 3). In our hands, children treated by this technique of pleural debridement have become afebrile more rapidly than those having the simple insertion of a tube . This procedure is not a formal decortication , because the fibrinous material is loose and easily removed with suction and forceps. Over the past 30 years, some authors have favored

The Journal of Thoracic and Cardiovascular Surgery

502 Raffensperger et at.

Fig. 3. A, Chest radiograph of a 4-month-old boy on the fifth day of illness showing extensive loculated staphylococcal empyema. B, Postoperative radiograph. Empyema completely drained through sixth rib. Patient afebrile on first postoperative day.

early operation in children when prompt resolution does not follow tube thoracostomy. 5-7 Early decortication recently has been recommended specifically for anaerobic empyema in children." In the past 15 to 20 years, a number of thoracic surgical groups have advocated early "decortication" in selected patients."?" Our results would indicate that early complete drainage and pleural debridement can hasten recovery in most cases. Modem effective antibiotics have nearly eliminated empyema as a cause for mortality in infants and children when combined with early adequate pleural drainage. We should choose drainage procedures which are humane and comfortable for the child and which decrease morbidity. The mini-thoracotomy is a step in this direction. REFERENCES

2

3

4

5

Jewett TC, Carberry DM, Neter E: Staphylococcal empyema in children. Ann Surg 153:447-452, 1961 Stiles QR, Lindesmith GG, Tucker BL, Meyer BW, Jones JC: Pleural empyema in children. Ann Thorac Surg 10:37-44, 1970 Sabiston DC, Hopkins EH, Cooke RE, Bennett IL: The surgical management of complications of staphylococcal pneumonia in infancy and childhood. J THORAC CARD10VASC SURG 38:421-434, 1959 Sandweiss DA, Hanson JC, Gosink BB, Moser KM: Ultrasound in diagnosis, localization, and treatment of loculated pleural empyema. Ann Intern Med 82:50-53, 1975 Hertzler JG, Miller AE, Tuttle WM: Present concepts in the treatment of empyema in children. Arch Surg 68:838-847, 1954

6 Middelkamp IN, Parderson ML, Burford TH: The changing pattern of empyema thoracis in pediatrics. J THoRAc CARDIOVASC SURG 47:165-173, 1964 7 Thomas MJ, Taylor FR, Sanger DW, Robicsek F: Decortication in the management of the complications of staphylococcal pneumonia in infants and children. J THORAC CARDIOVASC SURG 49:708-713, 1965 8 Kosloske AM, Cushing AH, Shuck JM: Early decortication for an aerobic empyema in children. J Pediatr Surg 15:422-429, 1980 9 Mayo P, McElvein RB: Early thoracotomy for pyogenic empyema. Ann Thorac Surg 2:649-657, 1966 IO Morin JE, Munro DD, MacLean LD: Early thoracotomy for empyema. J THORAC CARDIOVASC SURG 64:530-536, 1972 II Bryant LR, Chicklo JM, Crutcher R, Danielson GK, Malette WG, Trinkle JK: Management of thoracic empyema. J THORAC CARDIOVASC SURG 55:850-858, 1968 12 Fishman NH, Ellertson DG: Early pleural decortication for thoracic empyema in immunosuppressed patients. J THoRAc CARDIOVASC SURG 74:537-541, 1977 13 Mavroudis C, Symmonds JB, Minagi H, Thomas AN: Improved survival in management of empyema thoracis. J THORAC CARDIOVASC SURG 82:49-57, 1981

Discussion DR. G. HUGH LAWRENCE Portland. Ore.

The authors have reviewed very nicely the waxing and waning of enthusiam in the treatment of empyema since the early 1950s, when, with the virulent staphylococcal epidemics, the approach was fairly aggressive. However, as has been

Volume 84

Mini-thoracotomy for empyema

Number 4

50 3

October. 1982

pointed out by Stiles and others, since the advent of methicillin in 1962, decortication and open tube drainage are no longer considered as necessary, despite the awesome roentgenographic and gross surgical appearance that one encountered in those early days when decortication was still being performed. It soon became apparent that while one can develop a rather convincing argument based on appearances, as you have, maybe the restorative powers of the pleura should be allowed to produce an eventual clearing. In an effort to evaluate the added benefit of the minithoracotomy, which is less than a decortication but about the same as tube drainage and does require a general anesthetic, we have reviewed our experience in 21 children who have been treated over a similar period of time at the Oregon Health Sciences University Medical School Pediatric Service, largely by Dr. Campbell. We have largely used closed tube thoracostomy or thoracentesis alone with antibiotics. We used the more aggressive procedures only early in our experience. These empyemas were primarily postpneumonic, and there were surprisingly few staphylococcal cases as compared to our earlier experience. There were no deaths, and all patients had resolution of their clinical problem without recurrence, except for one patient with a congenital gamma globulin deficiency who had a recurrence 6 years ago, 2 years after the initial episode. We would conclude that at the present time there is little indication for a more extensive operative procedure than closed tube thoracostomy accompanied by appropriate antibiotic therapy. This will be accompanied by a subsequent resolution of the residual processes on an outpatient basis.

resected granuloma contained a small triangle of bony sequestrum that neatly replicated the deperiosteolized area demonstrated in Dr. Luck's slide. I would like to commend Dr. Luck for emphasizing that tube drainage of empyema is frequently ineffective and that surgical debridement of the semisolid exudate is a requirement for rapid recovery. DR. THOMAS M. HOLDER Kansas City. Mo.

In treating empyema, the object is to erradicate the bacteria and expand the lung. This requires that the pleural cavity be evacuated, and the type of treatment that is effective depends upon the stage of the disease. We have even seen a few patients who have been treated by thoracentesis and appropriate antibiotics. The next stage is a pleural cavity that can be evacuated by a drainage tube. After a period of time, however, the material becomes so fibrinous that tube drainage is ineffective. It is in this stage of the disease in which the approach advocated by Dr. Luck is very effective in evacuating the pleural cavity. We have treated a half dozen patients in the past 2 years with mini-thoracotomy. Results have been quite satisfactorysimilar to those reported by these authors. One technical point: We have not found it necessary to resect the rib and have used an intercostal incision at about the fifth intercostal space in the mid-axillary line. This approach provides full access to the pleural cavity and to the fissure in case there is an interlobar collection. Our experience supports the approach advocated by these authors.

DR. JURO WADA Tokyo. Japan

MR. MARK BRAIMBRIDGE

In doing a mini-thoracotomy, I would recommend that the rib cage be reconstructed by placing the resected rib back into its original position. Bony healing will be perfect, even in the presence of infection, as far as the wound is open. This would be useful in doing delicate neonatal or infant thoracic cage operations. I am positive in my statement on the basis of my more than 1,000 rib and perichondrial operations for funnel chest deformity.

London. England

DR. BENSON B. ROE San Francisco, Calif.

Dr. Luck's excellent presentation is marred only by a small but important flaw in her illustration of the rib resection technique. Some 40 years ago the late Dr. Edward D. Churchill emphasized that the rib whose segment is to be resected for empyema drainage should not be left with deperiosteolized triangles beyond the resected ends. To avoid that mistake, he recommended "T" incisions at both ends so as to leave complete periosteal covering on both transected ends. To emphasize the reality of this hazard, I will site a case which I reported some 25 years ago. In this patient, massive hemoptysis developed from a small granuloma in the region of a well-healed empyema drainage site from his childhood. The

We were equally dissatisfied with the necessity to do a thoracotomy in these children and also with the unsatisfactory nature of aspiration followed by tube drainage. Following the suggestion of Dr. Rosenfeldt, of Melbourne, Australia, we have been inserting a peritoneoscope (laparoscope) into the pleura in these children. With this device, one can break down all the loculi and suck out with a large sucker all the fibrin, if necessary picking some out with forceps and washing out the pleura with saline, a procedure that can be repeated by leaving in a small apical catheter. Clearance obtained in this way is as good as that obtained by removing a piece of rib. Then one is left with just an intercostal drain. We have been impressed with how this technique maximizes the effect of the intercostal drain without going to the length of removing a piece of rib. DR. L U C K (Closing) We would like to thank all of the discussers for their comments and for expanding our discussion. As Dr. Lawrence points out, closed tube drainage will suffice for definitive treatment in some children, but the length of hospitalization with toxicity and disability should be critically assessed be-

The Journal of

504 Raffensperger et al.

Thoracic and Cardiovascular Surgery

fore persisting with this form of treatment. Dr. Braimbridge's approach of drainage through thoracoscopy may be quite satisfactory in some of these children but will still require a general anesthetic in most cases. Perhaps the thoracoscopy could be combined with mini-thoracotomy when indicated. Dr. Roe, thank you for pointing out the error in our diagram-

matic representation of the technical procedure. You are absolutely right. [The diagram was corrected for publication.] As medical and surgical therapy improves in all respects, we have a continued obligation not only to effect a cure, but to do so with minimal morbidity and expense. Our approach to childhood empyema seeks these goals.