- Email: [email protected]
The changing spectrum of cystic pulmonary lesions requiring surgical resection in infants
J
THoRAc CARDIOVASC SURG
89:332-339, 1985
Original Communications
The changing spectrum of cystic pulmonary lesions requiring surgical resection in infants Congenital labor emphysema and cystic adenomatoid malformation have been the most common surgically treatable cystic lesions of the lung. With the successful treatment of newborn respiratory distress syndrome, an increased frequency of an acquired form of cystic disease, interstitial pulmonary emphysema, bas been observed. Some degree of interstitial pulmonary emphysema is relatively common, and during the years 1980 to 1983 this disease was diagnosed in 372 infants. Seven of these infants, all premature, underwent resection of relatively localized areas of persistent cystic interstitial pulmonary emphysema. Lobectomy or wedge resection was performed in five patients because of their inability to be weaned from the ventilator. A sixth patient with this disease underwent lobectomy for recurrent pneumothoraces. A seventh patient underwent lobectomy because the cystic interstitial pulmonary emphysema produced atelectasis and recurrent infections. All seven patients were extubated by the fourth postoperative day, have been discharged, and are showing respiratory improvement. Within the same period, four infants had congenital lobar emphysema and two had congenital cystic adenomatoid malformation. They were gestational age 36 weeks or older and, although respiratory distress was present to some degree in all six, only one was ventilator dependent at operation. All underwent lobectomy andone infant had a left upper lobe resection and right middle lobectomy on separate occasions for bilateral congenital cystic adenomatoid malformation. All patients with congenital lobar emphysema and congenital cystic adenomatoid malformation have been discharged and are doing weD. Our results suggest the foDowing conclusions: Persistent interstitial pulmonary emphysema is now the most common indication for pulmonary resection in the newborn period. The anatomic distribution of resected interstitial pulmonary emphysema is similar to that of congenital lobar emphysema. Although only a small fraction «2 %) of patients with interstitial pulmonary emphysema require operation, the development of relatively large discrete cystic areas that (1) significantly decrease effective lung volume and produce respirator dependence, (2) produce atelectasis and recurrent infections, or (3) lead to pneumothoraces may make pulmonary resection beneficial. FinaDy, despite the presence of generalized lung disease in patients with interstitial pulmonary emphysema, these patients can be expected to improve significantly after resection, and the long-term outcome is generaDy good.
Joseph R. Schneider, M.D., John A. St. Cyr, M.D., Theodore R. Thompson, M.D., Dana E. Johnson, M.D., Barbara A. Burke, M.D., and John E. Foker, M.D., Ph.D., Minneapolis. Minn. From the Departments of Surgery, Pediatrics, and Pathology, University of Minnesota Hospitals and Medical School, 420 Delaware SI. S.E., Minneapolis, Minn. 55455. Read at the Tenth Annual Meeting of The Western Thoracic Surgical Association, Maui, Hawaii, June 20-23, 1984. Address for reprints: Joseph R. Schneider, M.D., Box 311 Mayo, University Hospitals, 420 Delaware SI. S.E., Minneapolis, Minn. 55455.
332
h e spectrum of surgically treatable pulmonary disease in infants is changing. At one time infectious lesions (tuberculosis, bronchiectasis) were the most common indications for operation in children. Improving public health and antibiotics have been accompanied by a decline in these diseases. By the end of the past decade, the cystic pulmonary lesions that necessitated operation
Volume 89 Number 3
Cystic pulmonary lesions in infants
March. 1985
333
Fig. 1. Interstitial pulmonary emphysema. Left, Chestfilm takenshortly after admission, showing typical findings of infant respiratory distress syndrome. Right, X-ray film taken 3 days after admission is noteworthy for the development of a large, discrete cystic area in the right middle lobe.
Table I. Infants undergoing pulmonary resection at the University of Minnesota (1980 to 1983) Mean gestational age iwk]
Disease
IPE
Mean birth weight (gm}
29 (26-34)
1.314 (760-2,360)
CLE
4
39 (36-40)
3,838 (2,950-5,460)
CCAM
2
40 (40-41)
4,190 (3,400-4,980)
No. ventilator dependent at operation
Labels) resected
5
LUL (3) RML (2) LLL (I) RUL + RML (I) RUL (2) LUL (2) LUL (I) LUL + RML (I)
o
Legend: IPE, Interstitial pulmonary emphysema. CLE. Congenital lobar emphysema. CCAM. Congenital cystic adenomatoid malformations.
were predominantly congenital in origin. Congenital lobar emphysema and congenital cystic adenomatoid malformation constituted about 75% of these lesions.l' The clinical and pathological features of these lesions, as well as the indications for operation, are well described."? With the recent rapid advances in the treatment of infant respiratory distress syndrome (RDS), an acquired cystic lesion, interstitial pulmonary emphysema (IPE), has become a common diagnosis in neonatal intensive care units (NICUs). I.6 In contrast to the congenital cystic diseases, however, the incidence, indications, and results of operation in IPE have not yet been established. This report reviews the spectrum of lesions now encountered and, in particular, provides data on the surgical treatment of acquired cystic pulmonary lesions in infancy.
Patients During the years 1980 to 1983, 13 infants underwent lung resection for pulmonary cystic disease at the University of Minnesota Hospitals (Table I). Seven of these infants, four boys and three girls, underwent lobectomy or wedge resection for localized persistent IPE. The gestational ages of these infants ranged from 26 to 34 weeks and birth weights from 760 to 2,360 gm. The diagnosis of IPE was made in infants with RDS in whom lucent spaces developed within the lung parenchyma that were not present on the initial chest x-ray films taken shortly after birth (Fig. 1). Based on these criteria, this diagnosis was made in 372 infants in the NICU at the University of Minnesota Hospitals from 1980 to 1983. In almost all of these infants, the
The Journal of Thoracic and Cardiovascular Surgery
3 3 4 Schneider et al.
Fig. 2. Interstitial pulmonary emphysema. This infant with a large cystic area in the right middle lobe could not be weaned from the ventilator.
disease appeared to resolve with time. However, seven infants who developed relatively large localized areas of persistent IPE had continued respiratory difficulties and underwent resection of affected lung. Five infants with persistent IPE could not be weaned from the ventilator. These infants had shown general resolution of their hyaline membrane disease but had large residual discrete cystic areas within the lung fields. When the ventilator minute volume was reduced, they continued to show hypercarbia and respiratory acidosis despite adequate oxygenation. The large cysts contained considerable dead space, caused atelectasis of adjacent pulmonary tissue, and often produced significant mediastinal shifts (Fig. 2). These five infants had required an average of 29 (range 14 to 45) days of ventilator support before the operation. The operations performed included right upper and middle lobectomy in one infant, left upper lobectomy in three infants, and right middle lobectomy in one infant. Each of these five infants was able to be extubated by the fourth postoperative day and all show continued improvement in respiratory status. A sixth infant who had a large cystic area in the left lower lobe was extubated without an operation, but she had recurrent infection and persistent atelectasis in adjacent pulmonary tissue (Fig. 3). Consequently, she underwent
left lower lobectomy. She was extubated on the second postoperative day and is showing continued improvement in respiratory status. A seventh infant, who was extubated without an operation, had recurrent pneumothoraces. When this problem appeared to have resolved, he was discharged from the hospital. He had another pneumothorax and was readmitted to the hospital at the age of 2V2 months, at which time a right middle lobectomy was performed. This child was extubated on the first postoperative day and has had no recurrence of pneumothorax. He shows some developmental delay thought to be related to a cerebral intraventricular hemorrhage soon after birth, but is showing continued respiratory improvement. All seven of these patients have been discharged from the hospital. Four of the seven had residual bronchopulmonary dysplasia (BPD) severe enough to necessitate diuretics or bronchodilators at the time of discharge, but these medications have subsequently been discontinued in each case. During this same time period, six other infants underwent pulmonary resection for congential cystic lesions. All six of these infants were gestational age 36 weeks or older. Four infants, three boys and one girl, had congenital lobar emphysema, with two undergoing right upper lobectomy and two undergoing left upper lobectomy. Of the two remaining patients, both girls, one underwent left upper lobectomy and proved to have a type I cystic adenomatoid malformation. The second underwent left upper lobectomy at age 4 months and right middle lobectomy at age 3 years, with both specimens proving to be type II cystic adenomatoid malformations. Although all six of these infants had some degree of respiratory distress at operation, only one infant, who had congenital lobar emphysema, was ventilator dependent at operation, in contrast to five of the seven infants with IPE. All of these infants have been discharged from the hospital and are well. One infant with congenital lobar emphysema had an associated diaphragmatic duplication, but no other associated anomalies were recognized in any of the 13 patients. Pathology
The original pathological material was reviewed without knowledge of the original diagnosis. In all 13 cases the same diagnosis was again made. Pathological examination of the persistent IPE specimens showed irregularly shaped gas-filled cysts, often with bronchovascular bundles projecting into the lumen of the cysts. These cysts were interstitial and all included multinucleated giant cells "lining" the cysts (Fig. 4). Variable amounts of fibrotic tissue surrounded these cysts.
Volume 89 Number 3
Cystic pulmonary lesions in infants
March, 1985
335
Fig. 3. Interstitial pulmonary emphysema. Left. Chest film taken shortly after admission is typical for respiratory distress syndrome. Right. One month later, the chest film shows a large cyst with adjacent atelectasis of the left lower lobe. This child was successfully extubated but had multiple episodes of pneumonia and atelectasis.
The slides from the infants with persistent IPE were also carefully examined for abnormalities in the tissue adjacent to the cysts. Evolving BPD was present in five of seven cases but was variable in severity and was not distributed uniformly in any of the specimens. Atelectasis was also noted in four of seven cases. Active acute pneumonitis with thrombosis of pulmonary artery branches was seen in one case. In no case was persistent IPE the only pathological finding, and areas of totally normal lung could be found in only two cases. The four children with congenital lobar emphysema had large cystic areas, with the dilated alveoli characteristic of this disease (Fig. 5). The segmental bronchi were peculiar for the lack of organization of the cartilage, probably allowing bronchial collapse and air trapping. This deficiency of the cartilage is thought to be an important pathogenetic factor in many cases of congenitallobar emphysema. Interestingly, each specimen also had small areas of acute interstitial emphysema. Two cases of cystic adenomatoid malformation were encountered. In one case, classified as type I, there were relatively large cysts lined with respiratory epithelium, which produced atelectasis of surrounding lung tissue. In the other case, classified as type II, there were smaller, evenly spaced cysts lined by respiratory epithelium (Fig. 6). In addition, the same tissue diagnosis was made after a right middle lobectomy when the child was 3 years old.
The lesion was apparently small initially but subsequently grew and led to atelectasis and pneumonitis.
Discussion Persistent IPE now appears to be the most common indication for pulmonary resection in infants at centers with large NICUs. In these units, the success of treatment of infant RDS has been accompanied by an increased incidence of IPE. IPE does occur very rarely in term infants, but the positive end-expiratory pressure used to ventilate premature infants is associated with a much higher risk of IPE, pneumothorax, and pneumomediastinum." Our data appear to agree with this. During the period covered by this report, 633 neonates were admitted to the NICU with the diagnosis of RDS. IPE developed in 372 (59%) of these infants. This number was determined by chart review, and it is possible that a review of all the original chest films might identify an even larger number of cases. Although some degree of IPE appears to be common, very few patients require operation, and only seven, or less than 2%, of IPE patients eventually required pulmonary resection. The other reports of resection of persistent IPE in infancy have also stated that operation is rarely necessary, but the incidence was not given.':" IPE occurs most commonly in premature infants with RDS treated by artificial ventilation with positive end-
The Journal of
336
Schneider et ai.
Fig. 4. Interstitial pulmonary emphysema. This mediumsized interstitial cyst is typical of persistent interstitial pulmonary emphysema with fibrosis surrounding the cyst, a large vessel projecting into a cystic lumen (lower right) and the peculiar giant cells lining the cystic cavity (inset).
expiratory pressure. It is defined by the presence of air in the interstitial pulmonary tissue around the bronchi, blood vessels, within the interlobular septae, and subpleurally.':" In the acute stage of the disease, these gas-filled spaces are seen on chest radiographs as linear lucent areas, frequently radiating from the pulmonary hilum. This acute form of the disease is often associated with pneumothorax or pneumomediastinum. Most commonly IPE will resolve, but persistent IPE occurs in either a diffuse or predominantly localized form. When diffuse, the persistence of multiple gas-filled cysts in all lobes of the lungs may cause hyperexpansion of the thoracic cavity and depression of the diaphragm. It is usually associated with BPD and, therefore, chest radiographs may also show evidence of fibrosis and atelectasis typical of BPD. In localized persistent IPE, one or more lobes contain multiple cysts. These relatively large cysts, up to 3 em in diameter, tend to increase in size with time and cause atelectasis of adjacent lung tissue, mediastinal shift, depression of the hemidiaphragm, and recurrent pneumothoraces. I It is the localized form of the persistent disease that is potentially
Thoracic and Cardiovascular Surgery
Fig. 5. Congenital lobar emphysema. There is marked alveolar hyperexpansion due to air trapping.
amenable to surgical therapy, but the role of pulmonary resection in this disease is not yet well defined. Even the localized form of persistent IPE would be expected to have a background of generalized disease that might reduce the value of surgical resection. Certainly, the diffuse form of persistent IPE has been associated with clinical and pathological evidence of generalized BPD.I.6 Stocker and colleagues':" have stated that the localized form of the disease is less commonly associated with significant BPD. Pathological evaluation of the seven cases in this report suggests that there is some element of generalized lung disease even with localized persistent IPE. Nevertheless, the clinical results have been gratifying in this group of infants. The nonsurgical treatment of localized persistent IPE has been reported to include high inspired oxygen concentrations, II selective intubation of the bronchus serving the less affected lung," and high-frequency ventilation. I] However, the total number of patients included in these studies is small, and five infants survived from a total of seven patients. No firm conclusions can be drawn from these reports, but consideration can be given to these forms of treatment. High inspired oxygen levels could result in better gas
Volume 89 Number 3 March, 1985
absorption and regression of the cysts, although in patients with BPD this should not be given more than a short trial. Selective intubation of the contralateral bronchus was used in four of the five ventilatordependent infants in the present series but was unsuccessful in these patients. High-frequency ventilation of various types has been of some value in the treatment of diffuse IPE, but its worth in the localized form is unknown. Several reports of successful surgical treatment of patients with localized IPE after the failure of conservative management have appeared. Bauer and associates? reported the surgical treatment of three ventilatordependent premature infants, two of whom did not respond to selectivebronchial intubation. After resection (left upper, right lower, and left lower lobectomy), all patients were extubated within 8 days of the surgical procedure and were discharged from the hospital doing well. Similar good results were reported by Reyes and colleagues" in three patients who underwent lobectomy, one for persistent tension pneumothorax and two for inability to be weaned from the ventilator. Another surgical approach was taken by Levine, Trump, and Waterkotte," who performed a pleurotomy (multiple linear incisions of the visceral pleura overlying emphysematous blebs) on four patients, with good results in three patients and death of the fourth patient. These reports, together with the results with the seven patients in this series, establish that lobectomy can be done with very low morbidity and mortality, even in small and severely premature infants. Furthermore, despite a background of generalized disease, resection of localized persistent IPE has been successful.When the patients are carefully selected, significant and rapid clinical improvement can be anticipated. The spectrum of acquired cystic disease may be wider than encountered in this report. Cooney, Menke, and Allen10 reported on a series of five premature infants with RDS who underwent pulmonary resection. These patients all had infant RDS treated with artificial ventilation and positive end-expiratory pressure. The clinical presentation was similar to that with IPE with the development of pulmonary cystic disease and ventilator dependence. However, histologic examination of the resected specimens showed a lesion that appears to be distinct from IPE in four of the five cases. These patients had a more severe form .of acquired disease predominantly located in the right lower lobe and postulated by the authors to result from the trauma of endotracheal suctioning. These results suggest the following conclusions: Persistent IPE is now the most common indication for
Cystic pulmonary lesions in infants
337
Fig. 6. Congenital cystic adenomatoid malformation. Note the multiple small cysts, all lined by respiratory epithelium in the absence of bronchial or alveolar structures.
pulmonary resection in the newborn period. Although IPE appears to be common in premature infants treated with positive end-expiratory pressure, only a small fraction (less than 2%) of these patients will require resection. Persistent IPE necessitating resection is most often confined to the upper and middle lobes and is similar in distribution to congenital lobar emphysema. Even small and severely premature patients with relatively large discrete areas of localized persistent IPE, which produce significantly decreased effective lung volume, respirator dependence, cardiovascular compromise, atelectasis, recurrent infections, or recurrent pneumothoraces, may be benefited by pulmonary resection. Despite the presence of generalized lung disease in patients with localized persistent IPE, pulmonary resection can be expected to produce significant improvement in respiratory status, and the long-term outlook is generally good. REFERENCES Stocker JT, Drake RM, Madewell JE: Cystic and congenital lung disease in the newborn. Perspect Pediatr Pathol 4:93-154, 1978
3 3 8 Schneider et al.
2 Campbell DN, Lilly JR: The changing spectrum of pulmonary operations in infants and children. J THoRAc CARDIOVASC SURG 83:680-685, 1982 3 Buntain WL, Isaacs H, Payne VC, Lindesmith GG, Rosenkrantz JG: Lobar emphysema, cystic adenomatoid malformation, pulmonary sequestration and bronchogenic cysts in infancy and childhood: A clinical group. J Pediatr Surg 9:85-93, 1974 4 Stocker JT, Madewell JE, Drake RM: Congenital cystic adenomatoid malformation of the lung. Classification and morphologic spectrum. Hum Pathol 8: 155-171, 1977 5 Ostor AG, Fortune DW: Congenital cystic adenomatoid malformation of the lung. Am J Clin Pathol 70:595-604, 1978 6 Stocker JT, Madewell JE: Persistent interstitial pulmonary emphysema. Another complication of the respiratory distress syndrome. Pediatrics 59:847-857, 1977 7 Bauer CR, Brennan MJ, Doyle C, Poole CA: Surgical resection for pulmonary interstitial emphysema in the newborn infant. J Pediatr 93:656-661, 1978 8 Reyes HM, Kagan RJ, Rowlatt UF, Vidyasagai D: Pulmonary interstitial emphysema-surgical management. Report of three cases. J Pediatr Surg 15:266-269, 1980 9 Levine DH, Trump DS, Waterkotte G: Unilateral pulmonary interstitial emphysema. A surgical approach to treatment. Pediatrics 68:510-514, 1981 10 Cooney DR, Menke JA, Allen JE: "Acquired" lobar emphysema. A complication of respiratory distress in premature infants. J Pediatr Surg 12:897-904, 1977 II Leonidas JC, Hall RT, Rhodes PG: Conservative management of unilateral pulmonary interstitial emphysema under tension. J Pediatr 87:776-778, 1975 12 Brooks JG, Bustamante SA, Koops BL, Hilton S, Cooper D, Wesenberg RL, Simmons MA: Selective bronchial intubation for the treatment of several localized pulmonary interstitial emphysema in newborn infants. J Pediatr 91:648-652, 1977 13 Ng KPK, Easa D: Management of interstitial emphysema by high-frequency low positive-presure hand ventilation in the neonate. J Pediatr 95:117-118,1979 14 Leonidas JC, Bahn I, McCauley RGK: Persistent localized pulmonary intersitial emphysema and lymphangiectasia: A causal relationship? Pediatrics 64:165-171,1979
Discussion DR. ALBERT D. HALL Greenbrae. Ca/if.
The authors report that 2,076 newborn infants were treated in the NICU of the University of Minnesota Hospitals between 1980 and 1983. What percentage of those infants were premature and how many had hyaline membrane disease? The 18% incidence of acquired IPE might even be higher than stated. Clearly, there is a high incidence of acquired IPE associated with prolonged artificial ventilation with positive end-expiratory pressure.
The Journal of Thoracic and Cardiovascular Surgery
In this series there were seven infants, all premature, with acquired persistent, localized cystic lesions who were treated successfully by pulmonary resection. Five of these were ventilator-dependent infants. All of them did well and were extubated promptly postoperatively. During that same time there were six term infants from that unit who had successful pulmonary resection for congenital cystic lesions. There have been numerous papers about resection of congenital cysts in infants since Reinhoff's case report in 1934. In a report in 1974, the Los Angeles Children's Hospital group described their experience with 54 resections for relief of the effects of the space-occupying and air-trapping congenital cysts. No doubt the combined experience in this room would add many cases to the literature and would include such problems as massive cysts in neonates who have required treatment by emergency resection. Resection is a wellestablished and effective treatment for certain patients with congenital cystic disease of the lung. The authors have shown that when a large number of premature infants who have hyaline membrane disease are treated with present methods of ventilation, a small number will acquire localized pulmonary cysts which compromise ventilation. These cysts can be successfully resected using indications similar to those previously applied to the group with the congenital lesions. Dr. Schneider, perhaps it is not fair to call this acquired entity iatrogenic, but if it is, what has your group of neonatologists and pulmonologists done to focus on this problem, and have any changes been made that might alter the incidence? What is your definition of high-frequency ventilation? Are you talking about 70 to 100 cycles/min or are you talking about the super high-frequency method? What is the morbidity and mortality in the unoperated patients in this series? Were there patients who, in retrospect, would have benefited from resection? How do you differentiate between persistent localized and persistent diffuse cystic disease in selecting your patients for operation? Finally, are the proliferative BPD changes a reflection of pulmonary oxygen toxicity? Aren't those changes reversible and, having been reversed, will they then not allow more adequate ventilation of the involved segments? DR. SCHNEIDER (Closing) Of the 2,000+ premature infants admitted to the NICU in that 4 year period, 633 had hyaline membrane disease. That represents roughly half of the infants whom we termed premature (less than 36 weeks' gestation). The literature on surgical treatment of acquired cystic disease in infants is surprisingly scant. The largest series I could find prior to ours described only four cases. I have no doubt that others have much larger series, and it would be beneficial to learn of the results obtained by those who have seen more cases of this disease. We believe cystic pulmonary disease is related to high airway pressure, and we know of no way to prevent it. The general approach of the neonatologists to hyaline membrane
Volume 89 Number 3 March, 1985
disease is to minimize the maximum end-inspiratory and mean airway pressures. Beyond that, we have not been able to devise any other possible means of prevention. When I speak of high-frequency ventilation, I mean super high-frequency ventilation, such as 800 cycles/min, as described in the Scandinavian literature. The incidence of this disease in infants who are premature, but who do not require positive-pressure artificial ventilation, is less than 5%. Previous series have described the incidence of the disease in those requiring intubation and positive endexpiratory pressure to be at least 40%. In our series, the incidence was 59%. Obviously, it is best to minimize airway pressure and time on the ventilator in these infants. A retrospective review of all the patients with hyaline membrane disease would be a gargantuan task with this many infants. However, my co-author, Dr. Thompson, who is the director of the NICU, believes that there were no other infants who could have benefited from an operation during this time period, based on the results from these seven infants.
Cystic pulmonary lesions in infants
339
Without the benefit of autopsy, differentiating diffuse from localized disease cannot be done with certainty. Our criteria were (I) an obvious problem coming off the ventilator (which was mechanical), and (2) recurrent infection or pneumothorax in the face of what appeared to be localized cystic changes on the chest x-ray film. No doubt with careful inspection of these infants, if they had biopsy of the opposite lung, for instance, there would be some microscopic evidence of diffuse disease. I have little doubt that oxygen toxicity is the major cause of BPD in these infants. BPD does tend to resolve. Four of the seven infants who had resection were discharged on a regimen of bronchodilators or diuretics because of residual BPD. All four are now off all medication. Coming off the ventilator sooner may mean less BPD or earlier resolution. The evidence for that may be anecdotal and suggestive, but I certainly believe that the infants are going to do better if they come off the ventilator sooner, and I think surgery has helped us accomplish that.
THoRAc CARDIOVASC SURG
89:332-339, 1985
Original Communications
The changing spectrum of cystic pulmonary lesions requiring surgical resection in infants Congenital labor emphysema and cystic adenomatoid malformation have been the most common surgically treatable cystic lesions of the lung. With the successful treatment of newborn respiratory distress syndrome, an increased frequency of an acquired form of cystic disease, interstitial pulmonary emphysema, bas been observed. Some degree of interstitial pulmonary emphysema is relatively common, and during the years 1980 to 1983 this disease was diagnosed in 372 infants. Seven of these infants, all premature, underwent resection of relatively localized areas of persistent cystic interstitial pulmonary emphysema. Lobectomy or wedge resection was performed in five patients because of their inability to be weaned from the ventilator. A sixth patient with this disease underwent lobectomy for recurrent pneumothoraces. A seventh patient underwent lobectomy because the cystic interstitial pulmonary emphysema produced atelectasis and recurrent infections. All seven patients were extubated by the fourth postoperative day, have been discharged, and are showing respiratory improvement. Within the same period, four infants had congenital lobar emphysema and two had congenital cystic adenomatoid malformation. They were gestational age 36 weeks or older and, although respiratory distress was present to some degree in all six, only one was ventilator dependent at operation. All underwent lobectomy andone infant had a left upper lobe resection and right middle lobectomy on separate occasions for bilateral congenital cystic adenomatoid malformation. All patients with congenital lobar emphysema and congenital cystic adenomatoid malformation have been discharged and are doing weD. Our results suggest the foDowing conclusions: Persistent interstitial pulmonary emphysema is now the most common indication for pulmonary resection in the newborn period. The anatomic distribution of resected interstitial pulmonary emphysema is similar to that of congenital lobar emphysema. Although only a small fraction «2 %) of patients with interstitial pulmonary emphysema require operation, the development of relatively large discrete cystic areas that (1) significantly decrease effective lung volume and produce respirator dependence, (2) produce atelectasis and recurrent infections, or (3) lead to pneumothoraces may make pulmonary resection beneficial. FinaDy, despite the presence of generalized lung disease in patients with interstitial pulmonary emphysema, these patients can be expected to improve significantly after resection, and the long-term outcome is generaDy good.
Joseph R. Schneider, M.D., John A. St. Cyr, M.D., Theodore R. Thompson, M.D., Dana E. Johnson, M.D., Barbara A. Burke, M.D., and John E. Foker, M.D., Ph.D., Minneapolis. Minn. From the Departments of Surgery, Pediatrics, and Pathology, University of Minnesota Hospitals and Medical School, 420 Delaware SI. S.E., Minneapolis, Minn. 55455. Read at the Tenth Annual Meeting of The Western Thoracic Surgical Association, Maui, Hawaii, June 20-23, 1984. Address for reprints: Joseph R. Schneider, M.D., Box 311 Mayo, University Hospitals, 420 Delaware SI. S.E., Minneapolis, Minn. 55455.
332
h e spectrum of surgically treatable pulmonary disease in infants is changing. At one time infectious lesions (tuberculosis, bronchiectasis) were the most common indications for operation in children. Improving public health and antibiotics have been accompanied by a decline in these diseases. By the end of the past decade, the cystic pulmonary lesions that necessitated operation
Volume 89 Number 3
Cystic pulmonary lesions in infants
March. 1985
333
Fig. 1. Interstitial pulmonary emphysema. Left, Chestfilm takenshortly after admission, showing typical findings of infant respiratory distress syndrome. Right, X-ray film taken 3 days after admission is noteworthy for the development of a large, discrete cystic area in the right middle lobe.
Table I. Infants undergoing pulmonary resection at the University of Minnesota (1980 to 1983) Mean gestational age iwk]
Disease
IPE
Mean birth weight (gm}
29 (26-34)
1.314 (760-2,360)
CLE
4
39 (36-40)
3,838 (2,950-5,460)
CCAM
2
40 (40-41)
4,190 (3,400-4,980)
No. ventilator dependent at operation
Labels) resected
5
LUL (3) RML (2) LLL (I) RUL + RML (I) RUL (2) LUL (2) LUL (I) LUL + RML (I)
o
Legend: IPE, Interstitial pulmonary emphysema. CLE. Congenital lobar emphysema. CCAM. Congenital cystic adenomatoid malformations.
were predominantly congenital in origin. Congenital lobar emphysema and congenital cystic adenomatoid malformation constituted about 75% of these lesions.l' The clinical and pathological features of these lesions, as well as the indications for operation, are well described."? With the recent rapid advances in the treatment of infant respiratory distress syndrome (RDS), an acquired cystic lesion, interstitial pulmonary emphysema (IPE), has become a common diagnosis in neonatal intensive care units (NICUs). I.6 In contrast to the congenital cystic diseases, however, the incidence, indications, and results of operation in IPE have not yet been established. This report reviews the spectrum of lesions now encountered and, in particular, provides data on the surgical treatment of acquired cystic pulmonary lesions in infancy.
Patients During the years 1980 to 1983, 13 infants underwent lung resection for pulmonary cystic disease at the University of Minnesota Hospitals (Table I). Seven of these infants, four boys and three girls, underwent lobectomy or wedge resection for localized persistent IPE. The gestational ages of these infants ranged from 26 to 34 weeks and birth weights from 760 to 2,360 gm. The diagnosis of IPE was made in infants with RDS in whom lucent spaces developed within the lung parenchyma that were not present on the initial chest x-ray films taken shortly after birth (Fig. 1). Based on these criteria, this diagnosis was made in 372 infants in the NICU at the University of Minnesota Hospitals from 1980 to 1983. In almost all of these infants, the
The Journal of Thoracic and Cardiovascular Surgery
3 3 4 Schneider et al.
Fig. 2. Interstitial pulmonary emphysema. This infant with a large cystic area in the right middle lobe could not be weaned from the ventilator.
disease appeared to resolve with time. However, seven infants who developed relatively large localized areas of persistent IPE had continued respiratory difficulties and underwent resection of affected lung. Five infants with persistent IPE could not be weaned from the ventilator. These infants had shown general resolution of their hyaline membrane disease but had large residual discrete cystic areas within the lung fields. When the ventilator minute volume was reduced, they continued to show hypercarbia and respiratory acidosis despite adequate oxygenation. The large cysts contained considerable dead space, caused atelectasis of adjacent pulmonary tissue, and often produced significant mediastinal shifts (Fig. 2). These five infants had required an average of 29 (range 14 to 45) days of ventilator support before the operation. The operations performed included right upper and middle lobectomy in one infant, left upper lobectomy in three infants, and right middle lobectomy in one infant. Each of these five infants was able to be extubated by the fourth postoperative day and all show continued improvement in respiratory status. A sixth infant who had a large cystic area in the left lower lobe was extubated without an operation, but she had recurrent infection and persistent atelectasis in adjacent pulmonary tissue (Fig. 3). Consequently, she underwent
left lower lobectomy. She was extubated on the second postoperative day and is showing continued improvement in respiratory status. A seventh infant, who was extubated without an operation, had recurrent pneumothoraces. When this problem appeared to have resolved, he was discharged from the hospital. He had another pneumothorax and was readmitted to the hospital at the age of 2V2 months, at which time a right middle lobectomy was performed. This child was extubated on the first postoperative day and has had no recurrence of pneumothorax. He shows some developmental delay thought to be related to a cerebral intraventricular hemorrhage soon after birth, but is showing continued respiratory improvement. All seven of these patients have been discharged from the hospital. Four of the seven had residual bronchopulmonary dysplasia (BPD) severe enough to necessitate diuretics or bronchodilators at the time of discharge, but these medications have subsequently been discontinued in each case. During this same time period, six other infants underwent pulmonary resection for congential cystic lesions. All six of these infants were gestational age 36 weeks or older. Four infants, three boys and one girl, had congenital lobar emphysema, with two undergoing right upper lobectomy and two undergoing left upper lobectomy. Of the two remaining patients, both girls, one underwent left upper lobectomy and proved to have a type I cystic adenomatoid malformation. The second underwent left upper lobectomy at age 4 months and right middle lobectomy at age 3 years, with both specimens proving to be type II cystic adenomatoid malformations. Although all six of these infants had some degree of respiratory distress at operation, only one infant, who had congenital lobar emphysema, was ventilator dependent at operation, in contrast to five of the seven infants with IPE. All of these infants have been discharged from the hospital and are well. One infant with congenital lobar emphysema had an associated diaphragmatic duplication, but no other associated anomalies were recognized in any of the 13 patients. Pathology
The original pathological material was reviewed without knowledge of the original diagnosis. In all 13 cases the same diagnosis was again made. Pathological examination of the persistent IPE specimens showed irregularly shaped gas-filled cysts, often with bronchovascular bundles projecting into the lumen of the cysts. These cysts were interstitial and all included multinucleated giant cells "lining" the cysts (Fig. 4). Variable amounts of fibrotic tissue surrounded these cysts.
Volume 89 Number 3
Cystic pulmonary lesions in infants
March, 1985
335
Fig. 3. Interstitial pulmonary emphysema. Left. Chest film taken shortly after admission is typical for respiratory distress syndrome. Right. One month later, the chest film shows a large cyst with adjacent atelectasis of the left lower lobe. This child was successfully extubated but had multiple episodes of pneumonia and atelectasis.
The slides from the infants with persistent IPE were also carefully examined for abnormalities in the tissue adjacent to the cysts. Evolving BPD was present in five of seven cases but was variable in severity and was not distributed uniformly in any of the specimens. Atelectasis was also noted in four of seven cases. Active acute pneumonitis with thrombosis of pulmonary artery branches was seen in one case. In no case was persistent IPE the only pathological finding, and areas of totally normal lung could be found in only two cases. The four children with congenital lobar emphysema had large cystic areas, with the dilated alveoli characteristic of this disease (Fig. 5). The segmental bronchi were peculiar for the lack of organization of the cartilage, probably allowing bronchial collapse and air trapping. This deficiency of the cartilage is thought to be an important pathogenetic factor in many cases of congenitallobar emphysema. Interestingly, each specimen also had small areas of acute interstitial emphysema. Two cases of cystic adenomatoid malformation were encountered. In one case, classified as type I, there were relatively large cysts lined with respiratory epithelium, which produced atelectasis of surrounding lung tissue. In the other case, classified as type II, there were smaller, evenly spaced cysts lined by respiratory epithelium (Fig. 6). In addition, the same tissue diagnosis was made after a right middle lobectomy when the child was 3 years old.
The lesion was apparently small initially but subsequently grew and led to atelectasis and pneumonitis.
Discussion Persistent IPE now appears to be the most common indication for pulmonary resection in infants at centers with large NICUs. In these units, the success of treatment of infant RDS has been accompanied by an increased incidence of IPE. IPE does occur very rarely in term infants, but the positive end-expiratory pressure used to ventilate premature infants is associated with a much higher risk of IPE, pneumothorax, and pneumomediastinum." Our data appear to agree with this. During the period covered by this report, 633 neonates were admitted to the NICU with the diagnosis of RDS. IPE developed in 372 (59%) of these infants. This number was determined by chart review, and it is possible that a review of all the original chest films might identify an even larger number of cases. Although some degree of IPE appears to be common, very few patients require operation, and only seven, or less than 2%, of IPE patients eventually required pulmonary resection. The other reports of resection of persistent IPE in infancy have also stated that operation is rarely necessary, but the incidence was not given.':" IPE occurs most commonly in premature infants with RDS treated by artificial ventilation with positive end-
The Journal of
336
Schneider et ai.
Fig. 4. Interstitial pulmonary emphysema. This mediumsized interstitial cyst is typical of persistent interstitial pulmonary emphysema with fibrosis surrounding the cyst, a large vessel projecting into a cystic lumen (lower right) and the peculiar giant cells lining the cystic cavity (inset).
expiratory pressure. It is defined by the presence of air in the interstitial pulmonary tissue around the bronchi, blood vessels, within the interlobular septae, and subpleurally.':" In the acute stage of the disease, these gas-filled spaces are seen on chest radiographs as linear lucent areas, frequently radiating from the pulmonary hilum. This acute form of the disease is often associated with pneumothorax or pneumomediastinum. Most commonly IPE will resolve, but persistent IPE occurs in either a diffuse or predominantly localized form. When diffuse, the persistence of multiple gas-filled cysts in all lobes of the lungs may cause hyperexpansion of the thoracic cavity and depression of the diaphragm. It is usually associated with BPD and, therefore, chest radiographs may also show evidence of fibrosis and atelectasis typical of BPD. In localized persistent IPE, one or more lobes contain multiple cysts. These relatively large cysts, up to 3 em in diameter, tend to increase in size with time and cause atelectasis of adjacent lung tissue, mediastinal shift, depression of the hemidiaphragm, and recurrent pneumothoraces. I It is the localized form of the persistent disease that is potentially
Thoracic and Cardiovascular Surgery
Fig. 5. Congenital lobar emphysema. There is marked alveolar hyperexpansion due to air trapping.
amenable to surgical therapy, but the role of pulmonary resection in this disease is not yet well defined. Even the localized form of persistent IPE would be expected to have a background of generalized disease that might reduce the value of surgical resection. Certainly, the diffuse form of persistent IPE has been associated with clinical and pathological evidence of generalized BPD.I.6 Stocker and colleagues':" have stated that the localized form of the disease is less commonly associated with significant BPD. Pathological evaluation of the seven cases in this report suggests that there is some element of generalized lung disease even with localized persistent IPE. Nevertheless, the clinical results have been gratifying in this group of infants. The nonsurgical treatment of localized persistent IPE has been reported to include high inspired oxygen concentrations, II selective intubation of the bronchus serving the less affected lung," and high-frequency ventilation. I] However, the total number of patients included in these studies is small, and five infants survived from a total of seven patients. No firm conclusions can be drawn from these reports, but consideration can be given to these forms of treatment. High inspired oxygen levels could result in better gas
Volume 89 Number 3 March, 1985
absorption and regression of the cysts, although in patients with BPD this should not be given more than a short trial. Selective intubation of the contralateral bronchus was used in four of the five ventilatordependent infants in the present series but was unsuccessful in these patients. High-frequency ventilation of various types has been of some value in the treatment of diffuse IPE, but its worth in the localized form is unknown. Several reports of successful surgical treatment of patients with localized IPE after the failure of conservative management have appeared. Bauer and associates? reported the surgical treatment of three ventilatordependent premature infants, two of whom did not respond to selectivebronchial intubation. After resection (left upper, right lower, and left lower lobectomy), all patients were extubated within 8 days of the surgical procedure and were discharged from the hospital doing well. Similar good results were reported by Reyes and colleagues" in three patients who underwent lobectomy, one for persistent tension pneumothorax and two for inability to be weaned from the ventilator. Another surgical approach was taken by Levine, Trump, and Waterkotte," who performed a pleurotomy (multiple linear incisions of the visceral pleura overlying emphysematous blebs) on four patients, with good results in three patients and death of the fourth patient. These reports, together with the results with the seven patients in this series, establish that lobectomy can be done with very low morbidity and mortality, even in small and severely premature infants. Furthermore, despite a background of generalized disease, resection of localized persistent IPE has been successful.When the patients are carefully selected, significant and rapid clinical improvement can be anticipated. The spectrum of acquired cystic disease may be wider than encountered in this report. Cooney, Menke, and Allen10 reported on a series of five premature infants with RDS who underwent pulmonary resection. These patients all had infant RDS treated with artificial ventilation and positive end-expiratory pressure. The clinical presentation was similar to that with IPE with the development of pulmonary cystic disease and ventilator dependence. However, histologic examination of the resected specimens showed a lesion that appears to be distinct from IPE in four of the five cases. These patients had a more severe form .of acquired disease predominantly located in the right lower lobe and postulated by the authors to result from the trauma of endotracheal suctioning. These results suggest the following conclusions: Persistent IPE is now the most common indication for
Cystic pulmonary lesions in infants
337
Fig. 6. Congenital cystic adenomatoid malformation. Note the multiple small cysts, all lined by respiratory epithelium in the absence of bronchial or alveolar structures.
pulmonary resection in the newborn period. Although IPE appears to be common in premature infants treated with positive end-expiratory pressure, only a small fraction (less than 2%) of these patients will require resection. Persistent IPE necessitating resection is most often confined to the upper and middle lobes and is similar in distribution to congenital lobar emphysema. Even small and severely premature patients with relatively large discrete areas of localized persistent IPE, which produce significantly decreased effective lung volume, respirator dependence, cardiovascular compromise, atelectasis, recurrent infections, or recurrent pneumothoraces, may be benefited by pulmonary resection. Despite the presence of generalized lung disease in patients with localized persistent IPE, pulmonary resection can be expected to produce significant improvement in respiratory status, and the long-term outlook is generally good. REFERENCES Stocker JT, Drake RM, Madewell JE: Cystic and congenital lung disease in the newborn. Perspect Pediatr Pathol 4:93-154, 1978
3 3 8 Schneider et al.
2 Campbell DN, Lilly JR: The changing spectrum of pulmonary operations in infants and children. J THoRAc CARDIOVASC SURG 83:680-685, 1982 3 Buntain WL, Isaacs H, Payne VC, Lindesmith GG, Rosenkrantz JG: Lobar emphysema, cystic adenomatoid malformation, pulmonary sequestration and bronchogenic cysts in infancy and childhood: A clinical group. J Pediatr Surg 9:85-93, 1974 4 Stocker JT, Madewell JE, Drake RM: Congenital cystic adenomatoid malformation of the lung. Classification and morphologic spectrum. Hum Pathol 8: 155-171, 1977 5 Ostor AG, Fortune DW: Congenital cystic adenomatoid malformation of the lung. Am J Clin Pathol 70:595-604, 1978 6 Stocker JT, Madewell JE: Persistent interstitial pulmonary emphysema. Another complication of the respiratory distress syndrome. Pediatrics 59:847-857, 1977 7 Bauer CR, Brennan MJ, Doyle C, Poole CA: Surgical resection for pulmonary interstitial emphysema in the newborn infant. J Pediatr 93:656-661, 1978 8 Reyes HM, Kagan RJ, Rowlatt UF, Vidyasagai D: Pulmonary interstitial emphysema-surgical management. Report of three cases. J Pediatr Surg 15:266-269, 1980 9 Levine DH, Trump DS, Waterkotte G: Unilateral pulmonary interstitial emphysema. A surgical approach to treatment. Pediatrics 68:510-514, 1981 10 Cooney DR, Menke JA, Allen JE: "Acquired" lobar emphysema. A complication of respiratory distress in premature infants. J Pediatr Surg 12:897-904, 1977 II Leonidas JC, Hall RT, Rhodes PG: Conservative management of unilateral pulmonary interstitial emphysema under tension. J Pediatr 87:776-778, 1975 12 Brooks JG, Bustamante SA, Koops BL, Hilton S, Cooper D, Wesenberg RL, Simmons MA: Selective bronchial intubation for the treatment of several localized pulmonary interstitial emphysema in newborn infants. J Pediatr 91:648-652, 1977 13 Ng KPK, Easa D: Management of interstitial emphysema by high-frequency low positive-presure hand ventilation in the neonate. J Pediatr 95:117-118,1979 14 Leonidas JC, Bahn I, McCauley RGK: Persistent localized pulmonary intersitial emphysema and lymphangiectasia: A causal relationship? Pediatrics 64:165-171,1979
Discussion DR. ALBERT D. HALL Greenbrae. Ca/if.
The authors report that 2,076 newborn infants were treated in the NICU of the University of Minnesota Hospitals between 1980 and 1983. What percentage of those infants were premature and how many had hyaline membrane disease? The 18% incidence of acquired IPE might even be higher than stated. Clearly, there is a high incidence of acquired IPE associated with prolonged artificial ventilation with positive end-expiratory pressure.
The Journal of Thoracic and Cardiovascular Surgery
In this series there were seven infants, all premature, with acquired persistent, localized cystic lesions who were treated successfully by pulmonary resection. Five of these were ventilator-dependent infants. All of them did well and were extubated promptly postoperatively. During that same time there were six term infants from that unit who had successful pulmonary resection for congenital cystic lesions. There have been numerous papers about resection of congenital cysts in infants since Reinhoff's case report in 1934. In a report in 1974, the Los Angeles Children's Hospital group described their experience with 54 resections for relief of the effects of the space-occupying and air-trapping congenital cysts. No doubt the combined experience in this room would add many cases to the literature and would include such problems as massive cysts in neonates who have required treatment by emergency resection. Resection is a wellestablished and effective treatment for certain patients with congenital cystic disease of the lung. The authors have shown that when a large number of premature infants who have hyaline membrane disease are treated with present methods of ventilation, a small number will acquire localized pulmonary cysts which compromise ventilation. These cysts can be successfully resected using indications similar to those previously applied to the group with the congenital lesions. Dr. Schneider, perhaps it is not fair to call this acquired entity iatrogenic, but if it is, what has your group of neonatologists and pulmonologists done to focus on this problem, and have any changes been made that might alter the incidence? What is your definition of high-frequency ventilation? Are you talking about 70 to 100 cycles/min or are you talking about the super high-frequency method? What is the morbidity and mortality in the unoperated patients in this series? Were there patients who, in retrospect, would have benefited from resection? How do you differentiate between persistent localized and persistent diffuse cystic disease in selecting your patients for operation? Finally, are the proliferative BPD changes a reflection of pulmonary oxygen toxicity? Aren't those changes reversible and, having been reversed, will they then not allow more adequate ventilation of the involved segments? DR. SCHNEIDER (Closing) Of the 2,000+ premature infants admitted to the NICU in that 4 year period, 633 had hyaline membrane disease. That represents roughly half of the infants whom we termed premature (less than 36 weeks' gestation). The literature on surgical treatment of acquired cystic disease in infants is surprisingly scant. The largest series I could find prior to ours described only four cases. I have no doubt that others have much larger series, and it would be beneficial to learn of the results obtained by those who have seen more cases of this disease. We believe cystic pulmonary disease is related to high airway pressure, and we know of no way to prevent it. The general approach of the neonatologists to hyaline membrane
Volume 89 Number 3 March, 1985
disease is to minimize the maximum end-inspiratory and mean airway pressures. Beyond that, we have not been able to devise any other possible means of prevention. When I speak of high-frequency ventilation, I mean super high-frequency ventilation, such as 800 cycles/min, as described in the Scandinavian literature. The incidence of this disease in infants who are premature, but who do not require positive-pressure artificial ventilation, is less than 5%. Previous series have described the incidence of the disease in those requiring intubation and positive endexpiratory pressure to be at least 40%. In our series, the incidence was 59%. Obviously, it is best to minimize airway pressure and time on the ventilator in these infants. A retrospective review of all the patients with hyaline membrane disease would be a gargantuan task with this many infants. However, my co-author, Dr. Thompson, who is the director of the NICU, believes that there were no other infants who could have benefited from an operation during this time period, based on the results from these seven infants.
Cystic pulmonary lesions in infants
339
Without the benefit of autopsy, differentiating diffuse from localized disease cannot be done with certainty. Our criteria were (I) an obvious problem coming off the ventilator (which was mechanical), and (2) recurrent infection or pneumothorax in the face of what appeared to be localized cystic changes on the chest x-ray film. No doubt with careful inspection of these infants, if they had biopsy of the opposite lung, for instance, there would be some microscopic evidence of diffuse disease. I have little doubt that oxygen toxicity is the major cause of BPD in these infants. BPD does tend to resolve. Four of the seven infants who had resection were discharged on a regimen of bronchodilators or diuretics because of residual BPD. All four are now off all medication. Coming off the ventilator sooner may mean less BPD or earlier resolution. The evidence for that may be anecdotal and suggestive, but I certainly believe that the infants are going to do better if they come off the ventilator sooner, and I think surgery has helped us accomplish that.